Email alert
2015 Vol. 37, No. 3
column
Display Method:
2015, 37(3): 509-515.
doi: 10.11999/JEIT140799
Abstract:
The Concept Factorization (CF) algorithm can not take into account the label information and the multi-relationship of samples simultaneously. In this paper, a novel algorithm called Hyper-graph regularized Constrained Concept Factorization (HCCF) is proposed, which extracts the multi-geometry information of samples by constructing an undirected weighted hyper-graph Laplacian regularize term, hence overcomes the deficiency that traditional graph model expresses pair-wise relationship only. Meanwhile, HCCF takes full advantage of the label information of labeled samples as hard constraints, and it preserves label consistent in low-dimensional space. The objective function of HCCF is solved by the iterative multiplicative updating algorithm and its convergence is also proved. The experimental results on TDT2, Reuters, and PIE data sets show that the proposed approach achieves better clustering performance in terms of accuracy and normalized mutual information, and the effectiveness of the proposed approach is verified.
The Concept Factorization (CF) algorithm can not take into account the label information and the multi-relationship of samples simultaneously. In this paper, a novel algorithm called Hyper-graph regularized Constrained Concept Factorization (HCCF) is proposed, which extracts the multi-geometry information of samples by constructing an undirected weighted hyper-graph Laplacian regularize term, hence overcomes the deficiency that traditional graph model expresses pair-wise relationship only. Meanwhile, HCCF takes full advantage of the label information of labeled samples as hard constraints, and it preserves label consistent in low-dimensional space. The objective function of HCCF is solved by the iterative multiplicative updating algorithm and its convergence is also proved. The experimental results on TDT2, Reuters, and PIE data sets show that the proposed approach achieves better clustering performance in terms of accuracy and normalized mutual information, and the effectiveness of the proposed approach is verified.
2015, 37(3): 516-521.
doi: 10.11999/JEIT140931
Abstract:
In order to improve the robustness of the visual tracking algorithm based on sparse coding, the original sparse coding problem is decomposed into two sub sparse coding problems. And the size of the codebook is intensively increased while the computational cost is decreased. Furthermore, in order to decrease the number of the1-norm minimization, ridge regression is employed to exclude the intensive outlying particles via the reconstruction error. And the sparse representation of the particles with small reconstruction error is computed on the two subcodebooks. The high-dimension sparse representation is put into the classifier and the candidate with the biggest response is recognized as the target. The experiment results demonstrate that the robustness of the proposed algorithm is improved due to the employed Cartesian product of subcodebooks.
In order to improve the robustness of the visual tracking algorithm based on sparse coding, the original sparse coding problem is decomposed into two sub sparse coding problems. And the size of the codebook is intensively increased while the computational cost is decreased. Furthermore, in order to decrease the number of the1-norm minimization, ridge regression is employed to exclude the intensive outlying particles via the reconstruction error. And the sparse representation of the particles with small reconstruction error is computed on the two subcodebooks. The high-dimension sparse representation is put into the classifier and the candidate with the biggest response is recognized as the target. The experiment results demonstrate that the robustness of the proposed algorithm is improved due to the employed Cartesian product of subcodebooks.
2015, 37(3): 522-528.
doi: 10.11999/JEIT140481
Abstract:
Super-resolution image restoration methods based on Compressive Sensing (CS) generally adopt local sparse coding strategy. Such strategy encodes each image block independently, which easily induces artificial blocking effect. To overcome this problem, a super-resolution image restoration method based on nonlocal sparse coding is proposed. The nonlocal self-similarity of image is considered as a prior in the dictionary training and image coding processes, respectively. Specifically, the proposed algorithm trains the dictionary with interpolated low-resolution images, and calculates the weighted average local code of similar patches, in order to obtain the nonlocal sparse code of each image block. Numerical experiments suggest that the proposed algorithm has a good recovery performance, and is robust to image noise.
Super-resolution image restoration methods based on Compressive Sensing (CS) generally adopt local sparse coding strategy. Such strategy encodes each image block independently, which easily induces artificial blocking effect. To overcome this problem, a super-resolution image restoration method based on nonlocal sparse coding is proposed. The nonlocal self-similarity of image is considered as a prior in the dictionary training and image coding processes, respectively. Specifically, the proposed algorithm trains the dictionary with interpolated low-resolution images, and calculates the weighted average local code of similar patches, in order to obtain the nonlocal sparse code of each image block. Numerical experiments suggest that the proposed algorithm has a good recovery performance, and is robust to image noise.
2015, 37(3): 529-535.
doi: 10.11999/JEIT140374
Abstract:
A novel tracking algorithm is proposed that can work robustly in real-world scenarios, in order to overcome the problems associated with severe changes in pose, motion and occlusion. A discriminative model based on the superpixels and a generative model based on global color and gradient features are constructed respectively. Through combining these two models, the distinguishing and invariance of target appearance features description are increased. Furthermore, an update strategy based on sparse principal component analysis is proposed, which can reduce the redundancy of feature dictionary when it updates. A discrimination mechanism is added in the update process of discriminative model to alleviate the drift problem. The experimental results demonstrate that the proposed algorithm performs more stable and robustly compared with several state-of-the-art algorithms when dealing with complex situations such as pose variation, background interference, and occlusion.
A novel tracking algorithm is proposed that can work robustly in real-world scenarios, in order to overcome the problems associated with severe changes in pose, motion and occlusion. A discriminative model based on the superpixels and a generative model based on global color and gradient features are constructed respectively. Through combining these two models, the distinguishing and invariance of target appearance features description are increased. Furthermore, an update strategy based on sparse principal component analysis is proposed, which can reduce the redundancy of feature dictionary when it updates. A discrimination mechanism is added in the update process of discriminative model to alleviate the drift problem. The experimental results demonstrate that the proposed algorithm performs more stable and robustly compared with several state-of-the-art algorithms when dealing with complex situations such as pose variation, background interference, and occlusion.
Visual Tracking Based on Sparse Dense Structure Representation and Online Robust Dictionary Learning
2015, 37(3): 536-542.
doi: 10.11999/JEIT140507
Abstract:
The L1 trackers are robust to moderate occlusion. However, the L1 trackers are very computationally expensive and prone to model drift. To deal with these problems, firstly, a robust representation model is proposed based on sparse dense structure. The tracking robustness is improved by adding an L2 norm regularization on the coefficients associated with the target templates and L1 norm regularization on the coefficients associated with the trivial templates. To accelerate object tracking, a block coordinate optimization theory based fast numerical algorithm for the proposed representation model is designed via the ridge regression and the soft shrinkage operator. Secondly, to avoid model drift, an online robust dictionary learning algorithm is proposed for template update. Robust fast visual tracker is achieved via the proposed representation model and dictionary learning algorithm in particle filter framework. The experimental results on several challenging image sequences show that the proposed method has better performance than the state-of-the-art tracker.
The L1 trackers are robust to moderate occlusion. However, the L1 trackers are very computationally expensive and prone to model drift. To deal with these problems, firstly, a robust representation model is proposed based on sparse dense structure. The tracking robustness is improved by adding an L2 norm regularization on the coefficients associated with the target templates and L1 norm regularization on the coefficients associated with the trivial templates. To accelerate object tracking, a block coordinate optimization theory based fast numerical algorithm for the proposed representation model is designed via the ridge regression and the soft shrinkage operator. Secondly, to avoid model drift, an online robust dictionary learning algorithm is proposed for template update. Robust fast visual tracker is achieved via the proposed representation model and dictionary learning algorithm in particle filter framework. The experimental results on several challenging image sequences show that the proposed method has better performance than the state-of-the-art tracker.
2015, 37(3): 543-551.
doi: 10.11999/JEIT140433
Abstract:
Head pose estimation is an important evaluating indicator of human attention, which depends on many factors, such as illumination, noise, identification, occlusion and so on. In order to enhance estimation efficiency and accuracy, this paper presents tree-structure cascaded random forests to estimate head pose in different quality images. First, in order to eliminate the influence of different environment noise, combined texture features in random forests for positive facial patch classification are extracted, which will be the privileged inputs to estimate head pose. Second, a coarse-to-fine approach is proposed to estimate head pose both in the yaw and pitch, which is called tree-structure cascaded random forests. Third, an adaptive Gaussian mixture model is used to enhance discriminate vote energy in the tree distribution. This framework is evaluated in unconstrained environmental datasets. The experiments show that the proposed approach has a remarkable and robust performance in different quality images.
Head pose estimation is an important evaluating indicator of human attention, which depends on many factors, such as illumination, noise, identification, occlusion and so on. In order to enhance estimation efficiency and accuracy, this paper presents tree-structure cascaded random forests to estimate head pose in different quality images. First, in order to eliminate the influence of different environment noise, combined texture features in random forests for positive facial patch classification are extracted, which will be the privileged inputs to estimate head pose. Second, a coarse-to-fine approach is proposed to estimate head pose both in the yaw and pitch, which is called tree-structure cascaded random forests. Third, an adaptive Gaussian mixture model is used to enhance discriminate vote energy in the tree distribution. This framework is evaluated in unconstrained environmental datasets. The experiments show that the proposed approach has a remarkable and robust performance in different quality images.
2015, 37(3): 552-559.
doi: 10.11999/JEIT140505
Abstract:
Image denoising is an important issue in the field of image processing. This paper defines the three-dimensional axis distance on the basis of three-dimensional histogram, analyzes the distribution features of images three-dimensional axis distance under the influence of non-noise and salt-pepper noise situation, thus the denoising algorithm upon the three-dimensional axis distance is proposed. The proposed algorithm extends image edge at first and then detects noise by employing the three-dimensional axis distance. Finally, the noise is eliminated by excluding noise pixels median filter. The?results of comparisons among the proposed algorithm, median filter, adaptive switching median filter, efficient average filter, modified directional-weighted-median filter, and noise adaptive fuzzy switching median filter verify the effectiveness of the proposed algorithm.
Image denoising is an important issue in the field of image processing. This paper defines the three-dimensional axis distance on the basis of three-dimensional histogram, analyzes the distribution features of images three-dimensional axis distance under the influence of non-noise and salt-pepper noise situation, thus the denoising algorithm upon the three-dimensional axis distance is proposed. The proposed algorithm extends image edge at first and then detects noise by employing the three-dimensional axis distance. Finally, the noise is eliminated by excluding noise pixels median filter. The?results of comparisons among the proposed algorithm, median filter, adaptive switching median filter, efficient average filter, modified directional-weighted-median filter, and noise adaptive fuzzy switching median filter verify the effectiveness of the proposed algorithm.
2015, 37(3): 560-566.
doi: 10.11999/JEIT140587
Abstract:
Aiming at the limitation of existing methods for the medical ultrasound images despeckling, a novel quantum-inspired despeckling method based on the local entropy is proposed for the medical ultrasound images. Firstly, the log-transformed images are decomposed by the Dual-tree Complex Wavelet Transform (DTCWT), and the signal and speckle noise are modeled separately. Then, considering the normalized products of the local entropy of the real components extracted from coefficients and their parents, the adjustable parameter is obtained by the quantum inspired theory to adjust the probability of signal and noise. Finally, the modified bivariate shrinkage function is exploited to obtain the despeckled image. The experimental results show that the proposed method can preserve detail information effectively and reduce the speckle noise of medical ultrasound image at the same time.
Aiming at the limitation of existing methods for the medical ultrasound images despeckling, a novel quantum-inspired despeckling method based on the local entropy is proposed for the medical ultrasound images. Firstly, the log-transformed images are decomposed by the Dual-tree Complex Wavelet Transform (DTCWT), and the signal and speckle noise are modeled separately. Then, considering the normalized products of the local entropy of the real components extracted from coefficients and their parents, the adjustable parameter is obtained by the quantum inspired theory to adjust the probability of signal and noise. Finally, the modified bivariate shrinkage function is exploited to obtain the despeckled image. The experimental results show that the proposed method can preserve detail information effectively and reduce the speckle noise of medical ultrasound image at the same time.
2015, 37(3): 567-573.
doi: 10.11999/JEIT140419
Abstract:
1-Bit Compressive Sensing (CS) is an important branch of standard CS. The existing 1-Bit CS algorithm-Binary Iterative Hard Thresholding (BIHT) can perfectly recovery signals with high precision and consistency, which requires exact sparsity level in the recovery phase. Considering this problem, a new Sparsity Adaptive Binary Iterative Hard Thresholding (SABIHT) algorithm without prior information of the sparsity is proposed by modifying the BIHT algorithm. By using the adaptive process of automatically adjusting the hard threshold parameters and test conditions to estimate the sparsity, the proposed algorithm realizes accurate reconstruction and estimates the true supporting set of approximated signal. The analytical theory and simulation results show that the SABIHT algorithm recovers effectively the signals without prior information of signal sparsity and the reconstruction performance is similar to the BIHT algorithm.
1-Bit Compressive Sensing (CS) is an important branch of standard CS. The existing 1-Bit CS algorithm-Binary Iterative Hard Thresholding (BIHT) can perfectly recovery signals with high precision and consistency, which requires exact sparsity level in the recovery phase. Considering this problem, a new Sparsity Adaptive Binary Iterative Hard Thresholding (SABIHT) algorithm without prior information of the sparsity is proposed by modifying the BIHT algorithm. By using the adaptive process of automatically adjusting the hard threshold parameters and test conditions to estimate the sparsity, the proposed algorithm realizes accurate reconstruction and estimates the true supporting set of approximated signal. The analytical theory and simulation results show that the SABIHT algorithm recovers effectively the signals without prior information of signal sparsity and the reconstruction performance is similar to the BIHT algorithm.
2015, 37(3): 574-579.
doi: 10.11999/JEIT140668
Abstract:
The issue of single snapshot airspeed estimation is researched based on acoustic sensor array. According to the propagation property of acoustic waves in subsonic and supersonic air current, the output model of acoustic sensor array is constructed for a given measuring equipment. Then an airspeed estimation algorithm based on Sparse Covariance Matrix Iteration with a Single Snapshot (SCMISS) is presented. SCMISS has several unique features not shared by other sparse estimation methods: it does not require the user to make any difficult selection of regularization parameters, and it has lower computational complexity and better real-time. What is more, the proposed algorithm can be applied to both subsonic and supersonic circumstances with single snapshot measurement. Finally, a compact expression for the Cramr-Rao Bound (CRB) on the estimation error of airspeed is derived to evaluate the performance of the proposed algorithm. Simulations are implemented to show the effectiveness of SCMISS.
The issue of single snapshot airspeed estimation is researched based on acoustic sensor array. According to the propagation property of acoustic waves in subsonic and supersonic air current, the output model of acoustic sensor array is constructed for a given measuring equipment. Then an airspeed estimation algorithm based on Sparse Covariance Matrix Iteration with a Single Snapshot (SCMISS) is presented. SCMISS has several unique features not shared by other sparse estimation methods: it does not require the user to make any difficult selection of regularization parameters, and it has lower computational complexity and better real-time. What is more, the proposed algorithm can be applied to both subsonic and supersonic circumstances with single snapshot measurement. Finally, a compact expression for the Cramr-Rao Bound (CRB) on the estimation error of airspeed is derived to evaluate the performance of the proposed algorithm. Simulations are implemented to show the effectiveness of SCMISS.
2015, 37(3): 580-586.
doi: 10.11999/JEIT140613
Abstract:
Sparse microwave imaging is a novel theory that systematically introduces sparse signal processing to microwave imaging. Compared with conventional synthetic aperture radar imaging, sparse microwave imaging exhibits the advantage of better imagery quality and lower system complexity. Non-ambiguity reconstruction for sparse scene can be achieved on under-sampling raw data by means of sparse microwave imaging, which leads to total data amount reduction. The imagery quality of sparse microwave imaging depends on the recovery property of measurement matrix, which is affected by the sparse sampling strategy. This paper focuses on the problem of design the azimuth sparse sampling scheme. The connection between mutual coherence and recovery property of the measurement matrix is analyzed. A mutual coherence based criterion is then proposed and applied to optimize the existing azimuth sparse sampling scheme. Numerical results demonstrate the effectiveness of the proposed method and conclusions are discussed.
Sparse microwave imaging is a novel theory that systematically introduces sparse signal processing to microwave imaging. Compared with conventional synthetic aperture radar imaging, sparse microwave imaging exhibits the advantage of better imagery quality and lower system complexity. Non-ambiguity reconstruction for sparse scene can be achieved on under-sampling raw data by means of sparse microwave imaging, which leads to total data amount reduction. The imagery quality of sparse microwave imaging depends on the recovery property of measurement matrix, which is affected by the sparse sampling strategy. This paper focuses on the problem of design the azimuth sparse sampling scheme. The connection between mutual coherence and recovery property of the measurement matrix is analyzed. A mutual coherence based criterion is then proposed and applied to optimize the existing azimuth sparse sampling scheme. Numerical results demonstrate the effectiveness of the proposed method and conclusions are discussed.
2015, 37(3): 587-593.
doi: 10.11999/JEIT140509
Abstract:
Space debris often appears in the form of groups, and the radar echoes overlap each other along the range direction. Utilizing the block structure, a high resolution space debris imaging method of ISAR is proposed based on the block-sparse Compressed Sensing (CS). This method can get high resolution 1-D range profile of every debris based on the block-sparse CS with the characteristics of space debris, and obtain the ISAR image combined with the translation compensation and the Range Doppler (RD) algorithm. The simulation results illustrate that the proposed method can achieve high resolution ISAR image with less reconstruction error and iterative number compared with the non-structure CS method under limited measurements.
Space debris often appears in the form of groups, and the radar echoes overlap each other along the range direction. Utilizing the block structure, a high resolution space debris imaging method of ISAR is proposed based on the block-sparse Compressed Sensing (CS). This method can get high resolution 1-D range profile of every debris based on the block-sparse CS with the characteristics of space debris, and obtain the ISAR image combined with the translation compensation and the Range Doppler (RD) algorithm. The simulation results illustrate that the proposed method can achieve high resolution ISAR image with less reconstruction error and iterative number compared with the non-structure CS method under limited measurements.
2015, 37(3): 594-600.
doi: 10.11999/JEIT140713
Abstract:
SAR image of the ground target contains the target region formed by the scattered echoes of the target as well as the shadow area. However, the characteristics of the two areas are essentially different, therefore the traditional SAR image Automatic Target Recognition (ATR) methods use mainly target area information alone or shadow region only for recognition. This paper presents a joint sparse representation model by combining images of the shadow region and target region. By using the1\2 norm minimization method to solve the joint sparse representation model, the SAR image target recognition is achieved by minimizing the joint reconstruction error. Recognition results on Moving and Stationary Target Acquisition and Recognition (MSTAR) data sets show that the joint sparse representation model can effectively fuse the information within the target region and shadow region, and it has much better recognition performance than the methods using only the target or shadow area information of the image.
SAR image of the ground target contains the target region formed by the scattered echoes of the target as well as the shadow area. However, the characteristics of the two areas are essentially different, therefore the traditional SAR image Automatic Target Recognition (ATR) methods use mainly target area information alone or shadow region only for recognition. This paper presents a joint sparse representation model by combining images of the shadow region and target region. By using the1\2 norm minimization method to solve the joint sparse representation model, the SAR image target recognition is achieved by minimizing the joint reconstruction error. Recognition results on Moving and Stationary Target Acquisition and Recognition (MSTAR) data sets show that the joint sparse representation model can effectively fuse the information within the target region and shadow region, and it has much better recognition performance than the methods using only the target or shadow area information of the image.
2015, 37(3): 601-606.
doi: 10.11999/JEIT140707
Abstract:
For Synthetic Aperture Radar (SAR), the range ambiguity is one of the important factors causing significant deterioration of the imaging performance. Azimuth Phase Coding (APC) technique is an effective method to suppress the range ambiguity. However, since the suppression performance heavily depends on the system over-sampling rate, the APC technique could not have the same suppression performance for a multi-channel SAR system compared with a single-channel SAR system. This paper presents a novel method to suppress the range ambiguity for multi-channel SAR system based on the APC technique. First, the range ambiguity can be shifted in the azimuth frequency domain by using the APC technique, then by taking advantages of more phase centers of multiple channels, the Digital BeamForming (DBF) method can be used to filter out the range ambiguity and reconstruct the useful signal, thus most of the ambiguity components can be suppressed significantly. Finally, the simulation results validate the effectiveness of the proposed method.
For Synthetic Aperture Radar (SAR), the range ambiguity is one of the important factors causing significant deterioration of the imaging performance. Azimuth Phase Coding (APC) technique is an effective method to suppress the range ambiguity. However, since the suppression performance heavily depends on the system over-sampling rate, the APC technique could not have the same suppression performance for a multi-channel SAR system compared with a single-channel SAR system. This paper presents a novel method to suppress the range ambiguity for multi-channel SAR system based on the APC technique. First, the range ambiguity can be shifted in the azimuth frequency domain by using the APC technique, then by taking advantages of more phase centers of multiple channels, the Digital BeamForming (DBF) method can be used to filter out the range ambiguity and reconstruct the useful signal, thus most of the ambiguity components can be suppressed significantly. Finally, the simulation results validate the effectiveness of the proposed method.
2015, 37(3): 607-612.
doi: 10.11999/JEIT140659
Abstract:
The external calibration of shore-based sea clutter measurement radar faces the problems with influence of sea wave motion. This paper proposes a posture correction method based on a coordinate transformation. The proposed method which combines the radar signal acquisition process and radar antenna direction factor, can effectively solve the effect of sloshing on the accuracy of the external calibration. Using the external calibration trial data collected by L-band shore-based sea clutter measurement radar, the experimental results are analyzed and compared with posture correction improvement on the calibration accuracy. The results show that the proposed method can greatly improve the accuracy of the calibration and is important to provide technical support for universal adaptability of the real sea clutter data.
The external calibration of shore-based sea clutter measurement radar faces the problems with influence of sea wave motion. This paper proposes a posture correction method based on a coordinate transformation. The proposed method which combines the radar signal acquisition process and radar antenna direction factor, can effectively solve the effect of sloshing on the accuracy of the external calibration. Using the external calibration trial data collected by L-band shore-based sea clutter measurement radar, the experimental results are analyzed and compared with posture correction improvement on the calibration accuracy. The results show that the proposed method can greatly improve the accuracy of the calibration and is important to provide technical support for universal adaptability of the real sea clutter data.
2015, 37(3): 613-618.
doi: 10.11999/JEIT140626
Abstract:
The echo of the vehicle from the main lobe may contaminate the training samples of Space Time Adaptive Processing (STAP), which results in target self nulling effect, and therefore degrades the probability of detection. To mitigate this problem, this paper proposes a Knowledge Aided (KA) STAP which is based on the road network data to select the training samples. This study firstly estimates the radial velocity of vehicle to the radar; then the range-Doppler cells which may contain vehicle echo are obtained according to the velocity; in the following, this study distinguish whether the training samples contain vehicle echo according to the matching degree of the training samples with the steering vector of the main lobe and the clutter; finally, the samples containing vehicle echo are discarded when the covariance matrix for the STAP is estimated. The theory analysis and experimental results illustrate that the proposed method advances the output of signal to clutter plus noise ratio, and improves the performance of STAP in the road network environments.
The echo of the vehicle from the main lobe may contaminate the training samples of Space Time Adaptive Processing (STAP), which results in target self nulling effect, and therefore degrades the probability of detection. To mitigate this problem, this paper proposes a Knowledge Aided (KA) STAP which is based on the road network data to select the training samples. This study firstly estimates the radial velocity of vehicle to the radar; then the range-Doppler cells which may contain vehicle echo are obtained according to the velocity; in the following, this study distinguish whether the training samples contain vehicle echo according to the matching degree of the training samples with the steering vector of the main lobe and the clutter; finally, the samples containing vehicle echo are discarded when the covariance matrix for the STAP is estimated. The theory analysis and experimental results illustrate that the proposed method advances the output of signal to clutter plus noise ratio, and improves the performance of STAP in the road network environments.
2015, 37(3): 619-624.
doi: 10.11999/JEIT140678
Abstract:
In order to solve the problem that of High Frequency Surface Wave Radar (HFSWR) and Automatic Identification System (AIS) target point tracks fusion, a point tracks association algorithm using Jonker- Volgenant-Castanon (JVC) global optimal matching for different status is proposed. Firstly, the HFSWR and AIS target point tracks are divided into the quasi-static and dynamic data by the radial velocity. Then the radial velocity and spherical distance are selected as the feature parameters, and the different status data are respectively pre-associated by the radial velocity and spherical distance. Finally, the average of relative distance ratio is used to evaluate the effect of association. According to the selection of threshold parameter, the HFSWR and AIS point tracks are optimal associated with the JVC algorithm. The experimental results indicate that the proposed algorithm, in the condition of equal number point tracks associated, is superior to the Nearest Neighbor (NN) algorithm and Munkres association algorithm in the association accuracy, and the associate time is less than the NN algorithm and Munkres association. Moreover, three different time data gained from the target traits measured in nearly three years demonstrate that the feasibility and real-time of the proposed method.
In order to solve the problem that of High Frequency Surface Wave Radar (HFSWR) and Automatic Identification System (AIS) target point tracks fusion, a point tracks association algorithm using Jonker- Volgenant-Castanon (JVC) global optimal matching for different status is proposed. Firstly, the HFSWR and AIS target point tracks are divided into the quasi-static and dynamic data by the radial velocity. Then the radial velocity and spherical distance are selected as the feature parameters, and the different status data are respectively pre-associated by the radial velocity and spherical distance. Finally, the average of relative distance ratio is used to evaluate the effect of association. According to the selection of threshold parameter, the HFSWR and AIS point tracks are optimal associated with the JVC algorithm. The experimental results indicate that the proposed algorithm, in the condition of equal number point tracks associated, is superior to the Nearest Neighbor (NN) algorithm and Munkres association algorithm in the association accuracy, and the associate time is less than the NN algorithm and Munkres association. Moreover, three different time data gained from the target traits measured in nearly three years demonstrate that the feasibility and real-time of the proposed method.
2015, 37(3): 625-630.
doi: 10.11999/JEIT140584
Abstract:
This paper focuses on the polarimetric SAR tomography for forested areas based on compressive Multiple Signal Classification (MSC). First, full polarimetric SAR receives the reflected echo of the imaging area. Then, the signals from polarimetric channels are used to build multiple measurement vector model, and a wavelet basis is used in order to sparsely represent vertical structure. For achieving the measurement of forested area, the backscattering coefficients are reconstructed by Compressive Multiple Signal Classification (CMSC) algorithm. Simulated data from PolSARpro software and P-band data acquired by the E-SAR sensor of the German Aerospace Center validate that the method can effectively reduce the passes for SAR tomography and the probability of occurrence of spurious spikes under the same measurement accuracy.
This paper focuses on the polarimetric SAR tomography for forested areas based on compressive Multiple Signal Classification (MSC). First, full polarimetric SAR receives the reflected echo of the imaging area. Then, the signals from polarimetric channels are used to build multiple measurement vector model, and a wavelet basis is used in order to sparsely represent vertical structure. For achieving the measurement of forested area, the backscattering coefficients are reconstructed by Compressive Multiple Signal Classification (CMSC) algorithm. Simulated data from PolSARpro software and P-band data acquired by the E-SAR sensor of the German Aerospace Center validate that the method can effectively reduce the passes for SAR tomography and the probability of occurrence of spurious spikes under the same measurement accuracy.
2015, 37(3): 631-636.
doi: 10.11999/JEIT140697
Abstract:
When detecting low-attitude windshear with airborne weather radar, the real signals are usually covered with strong clutter. In this paper, a novel method of low-attitude windshear speed estimation based on Space-Time Adaptive Processing (STAP) is proposed to solve the above problem. The proposed method handles the range-dependence of clutter of airborne forward looking array with space-time interpretation theory to achieve the Independent and Identically Distributed (IID) samples used in the clutter covariance matrix estimation; then the space-time adaptive processor is constructed which is applicable to a distributed low-attitude windshear target to suppress clutter and accumulate windshear signal; finally the accurate estimation of wind speed is got. The experimental results show that the proposed method can achieve a superior clutter suppression performance and an accurate wind speed estimation in high clutter-to-noise ratio and low signal-to-noise ratio.
When detecting low-attitude windshear with airborne weather radar, the real signals are usually covered with strong clutter. In this paper, a novel method of low-attitude windshear speed estimation based on Space-Time Adaptive Processing (STAP) is proposed to solve the above problem. The proposed method handles the range-dependence of clutter of airborne forward looking array with space-time interpretation theory to achieve the Independent and Identically Distributed (IID) samples used in the clutter covariance matrix estimation; then the space-time adaptive processor is constructed which is applicable to a distributed low-attitude windshear target to suppress clutter and accumulate windshear signal; finally the accurate estimation of wind speed is got. The experimental results show that the proposed method can achieve a superior clutter suppression performance and an accurate wind speed estimation in high clutter-to-noise ratio and low signal-to-noise ratio.
2015, 37(3): 637-642.
doi: 10.11999/JEIT140824
Abstract:
In this paper, based on the mathematical model of conventional Binary Offset Carrier (BOC) modulation, a novel pseudo-code spread spectrum modulation scheme of Frequency Chirp Binary Offset Carrier (FC-BOC) is proposed. FC-BOC modulation scheme is constructed by introducing the binary offset carrier of frequency chirp instead of binary offset carrier of fixed frequency in the conventional BOC modulation mathematical model. Therefore, FC-BOC signal has a lot of characteristics inherited from BOC signal and several unique characteristics. Simulation and algorithm experiment show that the FC-BOC modulation scheme retains the primary advantages of BOC modulation, moreover, the FC-BOC signal acquires a narrow and sharp correlation main-peak with extremely suppressed side-peaks for avoiding multiple correlation peak ambiguity naturally, flat power spectral density function similar to band-pass white noise. FC-BOC modulation systems, either signal generating algorithms or signal receiving algorithms, have similar structure and equivalent complexity to conventional BOC modulation.
In this paper, based on the mathematical model of conventional Binary Offset Carrier (BOC) modulation, a novel pseudo-code spread spectrum modulation scheme of Frequency Chirp Binary Offset Carrier (FC-BOC) is proposed. FC-BOC modulation scheme is constructed by introducing the binary offset carrier of frequency chirp instead of binary offset carrier of fixed frequency in the conventional BOC modulation mathematical model. Therefore, FC-BOC signal has a lot of characteristics inherited from BOC signal and several unique characteristics. Simulation and algorithm experiment show that the FC-BOC modulation scheme retains the primary advantages of BOC modulation, moreover, the FC-BOC signal acquires a narrow and sharp correlation main-peak with extremely suppressed side-peaks for avoiding multiple correlation peak ambiguity naturally, flat power spectral density function similar to band-pass white noise. FC-BOC modulation systems, either signal generating algorithms or signal receiving algorithms, have similar structure and equivalent complexity to conventional BOC modulation.
2015, 37(3): 643-650.
doi: 10.11999/JEIT140575
Abstract:
Considering the limitations of phase unwrapping and measurement mechanism in the pulsed radar phase ranging technology, a three-dimension (3D) ranging mechanism based on distributed radar is proposed, and the observation matrix model of distributed radar 3D ranging is established. By using the sequential phase difference caused by the distributed antenna layout instead of the phase difference caused by pulse incubation, a distributed wide-band radar (3D) measurement based on observation matrix is implemented, and in?combination?with the target sparse distribution characteristics in the observed area, a wide-band pulse radar carrier measurement by phase based on CLEAN method is suggested. As shown by the simulation and analysis results, the distributed radar phase ranging mechanism and method in this paper not only solve the phase unwrapping problem in high dynamic and long distance measurement environment, but also acquire the three-dimension sequential images and position information of the observed target quickly. This proposed method provides a new technical support for the application of the radar ranging technology.
Considering the limitations of phase unwrapping and measurement mechanism in the pulsed radar phase ranging technology, a three-dimension (3D) ranging mechanism based on distributed radar is proposed, and the observation matrix model of distributed radar 3D ranging is established. By using the sequential phase difference caused by the distributed antenna layout instead of the phase difference caused by pulse incubation, a distributed wide-band radar (3D) measurement based on observation matrix is implemented, and in?combination?with the target sparse distribution characteristics in the observed area, a wide-band pulse radar carrier measurement by phase based on CLEAN method is suggested. As shown by the simulation and analysis results, the distributed radar phase ranging mechanism and method in this paper not only solve the phase unwrapping problem in high dynamic and long distance measurement environment, but also acquire the three-dimension sequential images and position information of the observed target quickly. This proposed method provides a new technical support for the application of the radar ranging technology.
2015, 37(3): 651-657.
doi: 10.11999/JEIT140466
Abstract:
A Track-Before-Detect (TBD) algorithm is presented to jointly detect and track multiple fluctuating targets under passive multistatic radar system based on Multi-target Multi-Bernoulli (MeMBer) filter. Because the amplitude likelihood is uncertain due to the unknown mean Signal-to-Noise Ratio (SNR) of fluctuating targets, firstly a uniform prior distribution is assumed for the mean SNR corresponding to the envelope output, and a likelihood function is marginalized over the range of possible values. Based on this approximated likelihood function, the fusion centre uses all the amplitude measurements from each receiver transmitter pair to update the predicted Bernoulli components. Simulations show that the proposed algorithm can jointly detect and track multiple fluctuating targets effectively, furthermore, the performance is similar to the situation of the known mean SNR when the value of the mean SNR is higher than 9 dB.
A Track-Before-Detect (TBD) algorithm is presented to jointly detect and track multiple fluctuating targets under passive multistatic radar system based on Multi-target Multi-Bernoulli (MeMBer) filter. Because the amplitude likelihood is uncertain due to the unknown mean Signal-to-Noise Ratio (SNR) of fluctuating targets, firstly a uniform prior distribution is assumed for the mean SNR corresponding to the envelope output, and a likelihood function is marginalized over the range of possible values. Based on this approximated likelihood function, the fusion centre uses all the amplitude measurements from each receiver transmitter pair to update the predicted Bernoulli components. Simulations show that the proposed algorithm can jointly detect and track multiple fluctuating targets effectively, furthermore, the performance is similar to the situation of the known mean SNR when the value of the mean SNR is higher than 9 dB.
2015, 37(3): 658-664.
doi: 10.11999/JEIT140679
Abstract:
The presence of dense deception jamming not only causes numerous false alarms but also raises the threshold of constant false alarm rate detector, which severely degrades the detection performance of airborne radar. To suppress the dense deception jamming, a moving target detection method based on data fitting is proposed. Firstly, a data basis matrix is constructed utilizing sample data adjacent to the cell under test, which is used to represent the test data in the form of least-square fitting. Simultaneously, an upper-bound constraint is adaptively calculated to protect target signals from being fitted. The proposed method can effectively suppress the dense deception jamming and significantly improve the moving target detection performance of airborne radar. The effectiveness of the proposed method is verified by the experimental results on measured radar data.
The presence of dense deception jamming not only causes numerous false alarms but also raises the threshold of constant false alarm rate detector, which severely degrades the detection performance of airborne radar. To suppress the dense deception jamming, a moving target detection method based on data fitting is proposed. Firstly, a data basis matrix is constructed utilizing sample data adjacent to the cell under test, which is used to represent the test data in the form of least-square fitting. Simultaneously, an upper-bound constraint is adaptively calculated to protect target signals from being fitted. The proposed method can effectively suppress the dense deception jamming and significantly improve the moving target detection performance of airborne radar. The effectiveness of the proposed method is verified by the experimental results on measured radar data.
2015, 37(3): 665-671.
doi: 10.11999/JEIT140672
Abstract:
A method based on the signal retransmission by the Geostationary Earth Orbit (GEO) comsat for the navigation satellite signal enhancement is proposed. At the signal transmitting side, utilizing the property of low power density of the Direct Sequence Spread Spectrum (DSSS), the power-controlled weak DSSS navigation signal, which is generated at the navigation center, can be modulated and retransmitted on the frequency band of the GEO comsat under the proper Bit Error Rate (BER) requirement of the communication signal. While at the receiving side, with the FREquency-SHift (FRESH) filter and a given adaptive signal cancellation algorithm, the strong communication signal can be well cancelled and thus the weak navigation enhanced signal is separated. The simulation results show that for the separated navigation enhanced signal, a relatively good acquisition and tracking performance can be got in its controlled power range, and this also verifies the effectiveness of the proposed method.
A method based on the signal retransmission by the Geostationary Earth Orbit (GEO) comsat for the navigation satellite signal enhancement is proposed. At the signal transmitting side, utilizing the property of low power density of the Direct Sequence Spread Spectrum (DSSS), the power-controlled weak DSSS navigation signal, which is generated at the navigation center, can be modulated and retransmitted on the frequency band of the GEO comsat under the proper Bit Error Rate (BER) requirement of the communication signal. While at the receiving side, with the FREquency-SHift (FRESH) filter and a given adaptive signal cancellation algorithm, the strong communication signal can be well cancelled and thus the weak navigation enhanced signal is separated. The simulation results show that for the separated navigation enhanced signal, a relatively good acquisition and tracking performance can be got in its controlled power range, and this also verifies the effectiveness of the proposed method.
2015, 37(3): 672-678.
doi: 10.11999/JEIT140371
Abstract:
When existing multi-objective evolutionary clustering algorithms is applied to image segmentation, it can not obtain satisfactory segmentation performance on an image corrupted by noise due to no consideration of any spatial information derived from the image. Based on the complementarity of the local spatial information and the non local spatial information of the image, these two kinds of spatial information are introduced into a cluster validity function, and a novel objective function with complementary spatial information is constructed, and then a multi-objective evolutionary clustering algorithm with complementary spatial information for image segmentation is proposed. In order to reduce human intervention, the variable string length real coded technique is adopted to determine automatically the number of clusters during the evolving process. Natural image segmentation experiments show that the proposed method not only can obtain satisfactory segmentation performance on noisy images, but also can be suitable for many types of noisy images.
When existing multi-objective evolutionary clustering algorithms is applied to image segmentation, it can not obtain satisfactory segmentation performance on an image corrupted by noise due to no consideration of any spatial information derived from the image. Based on the complementarity of the local spatial information and the non local spatial information of the image, these two kinds of spatial information are introduced into a cluster validity function, and a novel objective function with complementary spatial information is constructed, and then a multi-objective evolutionary clustering algorithm with complementary spatial information for image segmentation is proposed. In order to reduce human intervention, the variable string length real coded technique is adopted to determine automatically the number of clusters during the evolving process. Natural image segmentation experiments show that the proposed method not only can obtain satisfactory segmentation performance on noisy images, but also can be suitable for many types of noisy images.
2015, 37(3): 686-692.
doi: 10.11999/JEIT140491
Abstract:
The current resource scheduling algorithms can not offer promise for the fairness of shared resources based on research on the resource scheduling algorithms in the system virtualization tools represented by Xen. This paper proposes a Historical Forwarding Overhead Based the Resource Scheduling Algorithm (HFOB_RSA) to ensure that router instances occupy determinate physical resources including I/Os that should be proportional to actual requirements. This algorithm can determine the scheduling priority of router instances by calculating the last several forwarding overhead of each router instance to make those router instances with low real throughput have an opportunity for being scheduled. Meanwhile, this algorithm also sets the value of processing delay in priority formula to provide support for those delay-sensitive router instances and determines whether or not to discard the non-disposed packets in advance by forecasting their processing time to avoid the unwanted overhead. The experimental results prove that the HFOB_RSA has superiority over Credit algorithm in terms of the fairness of shared resources in the virtual router. And HFOB_RSA also is able to provide support for the delay-sensitive router instances.
The current resource scheduling algorithms can not offer promise for the fairness of shared resources based on research on the resource scheduling algorithms in the system virtualization tools represented by Xen. This paper proposes a Historical Forwarding Overhead Based the Resource Scheduling Algorithm (HFOB_RSA) to ensure that router instances occupy determinate physical resources including I/Os that should be proportional to actual requirements. This algorithm can determine the scheduling priority of router instances by calculating the last several forwarding overhead of each router instance to make those router instances with low real throughput have an opportunity for being scheduled. Meanwhile, this algorithm also sets the value of processing delay in priority formula to provide support for those delay-sensitive router instances and determines whether or not to discard the non-disposed packets in advance by forecasting their processing time to avoid the unwanted overhead. The experimental results prove that the HFOB_RSA has superiority over Credit algorithm in terms of the fairness of shared resources in the virtual router. And HFOB_RSA also is able to provide support for the delay-sensitive router instances.
2015, 37(3): 693-699.
doi: 10.11999/JEIT140557
Abstract:
Reliability problems on Exchanged Hypercube interconnection network (EH(s, t))regard as one of important candidates of network models in large-scale processor systems are concerned by people. The extra connectivity, which is an important measure in evaluating the reliability, is utilized to analyze the reliability of exchanged hypercube interconnection network. Then the 2-extra vertex connectivity(k2(EH(s, t))) and 2-extra edge connectivity(2(EH(s, t))) of exchanged hypercube interconnection network are obtained. The conclusions are thatk2(EH(s, t))= 3s-2 for ts2; and 2(EH(s, t))=3s-1 for ts3 The analysis shows that the 2-extra connectivity is much superior to the traditional connectivity in evaluating the reliability of exchanged hypercube interconnection network.
Reliability problems on Exchanged Hypercube interconnection network (EH(s, t))regard as one of important candidates of network models in large-scale processor systems are concerned by people. The extra connectivity, which is an important measure in evaluating the reliability, is utilized to analyze the reliability of exchanged hypercube interconnection network. Then the 2-extra vertex connectivity(k2(EH(s, t))) and 2-extra edge connectivity(2(EH(s, t))) of exchanged hypercube interconnection network are obtained. The conclusions are thatk2(EH(s, t))= 3s-2 for ts2; and 2(EH(s, t))=3s-1 for ts3 The analysis shows that the 2-extra connectivity is much superior to the traditional connectivity in evaluating the reliability of exchanged hypercube interconnection network.
2015, 37(3): 700-707.
doi: 10.11999/JEIT140527
Abstract:
How to efficiently utilize the temporary cached replicas poses challenges to the retrieval process of Content Centric Networking (CCN). Inspired by the idea of data fields, a hybrid routing scheme based on differentiated cache advertisement is proposed. In the scheme, depending on the content popularity and resident probability, the differentiated advertisement strategy is performed to choose the advertised content and calculate the attracting potential. When a retrieve is requested, in order to realize global routing and local response to the content request, the globally oriented potential field and locally attracting potential field are constructed for the stable original content source and volatile cached copies, respectively. The simulation results show that the scheme can decrease the request latency, increase the cache hit ratio, while improving the overall performance of content delivery with a small amount of additional overhead.
How to efficiently utilize the temporary cached replicas poses challenges to the retrieval process of Content Centric Networking (CCN). Inspired by the idea of data fields, a hybrid routing scheme based on differentiated cache advertisement is proposed. In the scheme, depending on the content popularity and resident probability, the differentiated advertisement strategy is performed to choose the advertised content and calculate the attracting potential. When a retrieve is requested, in order to realize global routing and local response to the content request, the globally oriented potential field and locally attracting potential field are constructed for the stable original content source and volatile cached copies, respectively. The simulation results show that the scheme can decrease the request latency, increase the cache hit ratio, while improving the overall performance of content delivery with a small amount of additional overhead.
2015, 37(3): 708-714.
doi: 10.11999/JEIT140408
Abstract:
In Wireless Sensor Network (WSN), transferring all training samples distributed across different nodes to a centralized fusion center for training Support Vector Machine (SVM) significantly increases the communication overhead and energy consumption. Therefore, this paper studies the distributed training approach for linear SVM through the collaboration of neighboring nodes within the networks. First, the centralized linear SVM problem is cast as the solution of coupled decentralized convex optimization sub-problems with consensus constraints on the classifier parameters. Second, the distributed linear SVM problem is solved and derived using the augmented Lagrange multipliers method, and a novel distributed training algorithm, called Average Consensus based Distributed Supported Vector Machine (AC-DSVM), is proposed. To decrease the communication overhead of global average consensus, an improved distributed training algorithm, named Once Average Consensus based Distributed Supported Vector Machine (1-AC-DSVM), is presented, which is only based on once global average consensus. Simulation results show that compared with existing algorithms, AC-DSVM has slightly higher iterations and data traffic, but can converge to the centralized training results; 1-AC-DSVM not only has better convergence, but also has remarkable advantage in convergence speed and data traffic.
In Wireless Sensor Network (WSN), transferring all training samples distributed across different nodes to a centralized fusion center for training Support Vector Machine (SVM) significantly increases the communication overhead and energy consumption. Therefore, this paper studies the distributed training approach for linear SVM through the collaboration of neighboring nodes within the networks. First, the centralized linear SVM problem is cast as the solution of coupled decentralized convex optimization sub-problems with consensus constraints on the classifier parameters. Second, the distributed linear SVM problem is solved and derived using the augmented Lagrange multipliers method, and a novel distributed training algorithm, called Average Consensus based Distributed Supported Vector Machine (AC-DSVM), is proposed. To decrease the communication overhead of global average consensus, an improved distributed training algorithm, named Once Average Consensus based Distributed Supported Vector Machine (1-AC-DSVM), is presented, which is only based on once global average consensus. Simulation results show that compared with existing algorithms, AC-DSVM has slightly higher iterations and data traffic, but can converge to the centralized training results; 1-AC-DSVM not only has better convergence, but also has remarkable advantage in convergence speed and data traffic.
2015, 37(3): 715-720.
doi: 10.11999/JEIT140704
Abstract:
A path based Integer Linear Programming (ILP) method is proposed to optimize the network energy consumption under the bandwidth constrained transparent IP over WDM network model. Compared with the link based ILP method, this method can provide more lightpath combinations in the optical layer. The simulation results show that the path based ILP method can select the better lightpath combinations than the link based ILP method, and achieve lower network energy consumption.
A path based Integer Linear Programming (ILP) method is proposed to optimize the network energy consumption under the bandwidth constrained transparent IP over WDM network model. Compared with the link based ILP method, this method can provide more lightpath combinations in the optical layer. The simulation results show that the path based ILP method can select the better lightpath combinations than the link based ILP method, and achieve lower network energy consumption.
2015, 37(3): 721-726.
doi: 10.11999/JEIT140708
Abstract:
Partial-Band Noise Jamming (PBNJ) is one main type of narrow-band jamming, it has a huge impact on the performance of communication systems. The minimum resolution of jamming bandwidth of the conventional PBNJ model is the Frequency-Hopping (FH) sub-band bandwidth (Multiple Frequency Shift Keying (MFSK) signal bandwidth) in the FH/MFSK system. However, it is not always reasonable, thus a new PBNJ model, whose minimum resolution of jamming bandwidth can accurate to 1/M of the MFSK signal bandwidth is studied. In this paper, the closed-form expressions of Bit Error Rate (BER) performance under the new PBNJ model over Rician fading channel are derived and validated by computer simulations. The theoretical and simulation results show that the BER performance difference between the new and conventional PBNJ models is larger for smaller M,Nh,.
Partial-Band Noise Jamming (PBNJ) is one main type of narrow-band jamming, it has a huge impact on the performance of communication systems. The minimum resolution of jamming bandwidth of the conventional PBNJ model is the Frequency-Hopping (FH) sub-band bandwidth (Multiple Frequency Shift Keying (MFSK) signal bandwidth) in the FH/MFSK system. However, it is not always reasonable, thus a new PBNJ model, whose minimum resolution of jamming bandwidth can accurate to 1/M of the MFSK signal bandwidth is studied. In this paper, the closed-form expressions of Bit Error Rate (BER) performance under the new PBNJ model over Rician fading channel are derived and validated by computer simulations. The theoretical and simulation results show that the BER performance difference between the new and conventional PBNJ models is larger for smaller M,Nh,.
2015, 37(3): 727-732.
doi: 10.11999/JEIT140725
Abstract:
In order to apply Visible Light Communication (VLC) to indoor positioning, based on the Received Signal Strength (RSS) positioning technology, a scheme namely MLED-RSS positioning algorithm, utilizing multiple LED transmitters to achieve indoor positioning is proposed in this paper. In the proposed scheme, the impact of topology on positioning performance is fully considered, three LEDs are reasonably selected from the multiple LEDs deployed in the room as transmitted nodes to provide position coordinates, and then the improved trilateration method is used to obtain the target location information. MLED-RSS positioning algorithm can effectively solve the block or shadow effect existing in indoor visible location. Simulation results show that the MLED-RSS positioning algorithm can achieve high localization accuracy.
In order to apply Visible Light Communication (VLC) to indoor positioning, based on the Received Signal Strength (RSS) positioning technology, a scheme namely MLED-RSS positioning algorithm, utilizing multiple LED transmitters to achieve indoor positioning is proposed in this paper. In the proposed scheme, the impact of topology on positioning performance is fully considered, three LEDs are reasonably selected from the multiple LEDs deployed in the room as transmitted nodes to provide position coordinates, and then the improved trilateration method is used to obtain the target location information. MLED-RSS positioning algorithm can effectively solve the block or shadow effect existing in indoor visible location. Simulation results show that the MLED-RSS positioning algorithm can achieve high localization accuracy.
2015, 37(3): 733-738.
doi: 10.11999/JEIT140665
Abstract:
The existing methods of Doppler estimation in UnderWater Acoustic (UWA) communication generally assume that the Doppler factor of each path is equal in multipath channel. But when the Doppler factors are different from each other, such methods can not estimate the Doppler factor correctly, resulting in significant compensation error. By analyzing the sparsity of the UWA channel, a novel method of non-uniform Doppler estimation is proposed, which is based on the sparse representation theory. By using the over-complete dictionary combining UWA multipath channel and non-uniform Doppler sparsity, the method transforms Doppler estimation of each path into sparse solution reconstruction, achieving accurate estimation of non-uniform Doppler factor. Simulation results show that the proposed method can estimate not only the quite different Doppler factors but also the large Doppler frequency offset. Therefore, it is especially important for Doppler estimation in high-speed mobile UWA communication.
The existing methods of Doppler estimation in UnderWater Acoustic (UWA) communication generally assume that the Doppler factor of each path is equal in multipath channel. But when the Doppler factors are different from each other, such methods can not estimate the Doppler factor correctly, resulting in significant compensation error. By analyzing the sparsity of the UWA channel, a novel method of non-uniform Doppler estimation is proposed, which is based on the sparse representation theory. By using the over-complete dictionary combining UWA multipath channel and non-uniform Doppler sparsity, the method transforms Doppler estimation of each path into sparse solution reconstruction, achieving accurate estimation of non-uniform Doppler factor. Simulation results show that the proposed method can estimate not only the quite different Doppler factors but also the large Doppler frequency offset. Therefore, it is especially important for Doppler estimation in high-speed mobile UWA communication.
2015, 37(3): 739-745.
doi: 10.11999/JEIT140486
Abstract:
Based on analyzing the dispersion characteristics and global band gaps for general two-dimensional Photonic Band Gap (PBG) structures formed by triangular arrays of metal posts for TE modes, two methods are proposed to solve the problem of the high Q value in PBG cavity. The result shows that the Q value can be effectively controlled by loading media structures or inserting dielectric perturbation in PBG cavity. In consideration of mode competition in PBG cavity, the mode selection is calculated respectively for the two methods. It is shown that the two methods both can control the Q value without reducing the mode selection or changing the electric field distribution of TE04 mode. In addition, two competing modes are cleared in the method of dielectric perturbation, which improves the mode selection at the operation frequency of TE04 mode.
Based on analyzing the dispersion characteristics and global band gaps for general two-dimensional Photonic Band Gap (PBG) structures formed by triangular arrays of metal posts for TE modes, two methods are proposed to solve the problem of the high Q value in PBG cavity. The result shows that the Q value can be effectively controlled by loading media structures or inserting dielectric perturbation in PBG cavity. In consideration of mode competition in PBG cavity, the mode selection is calculated respectively for the two methods. It is shown that the two methods both can control the Q value without reducing the mode selection or changing the electric field distribution of TE04 mode. In addition, two competing modes are cleared in the method of dielectric perturbation, which improves the mode selection at the operation frequency of TE04 mode.
2015, 37(3): 746-752.
doi: 10.11999/JEIT140518
Abstract:
In order to solve the contradiction between the efficiency and accuracy as using the existing methods in the processing of the singularity of electromagnetic field near the microstrip line edge, in this paper, a microstrip line edge singularity processing technology based on the Locally One Dimensional Finite Difference Time Domain (LOD-FDTD), and combined with distribution function of electromagnetic field near microstrip line edge is proposed. The algorithm can handle conductors embedded in the grid area of more than 1/2 by the coordinate transformation, thus having wide applicability. Compared with the existing processing technology, the proposed algorithm in this paper has higher calculation accuracy, when the time step size is less than or equal to 5 times of the Courant-Friedrichs-Lewy (CFL) condition allowed. And compared with general LOD-FDTD, the proposed algorithm by introducing distribution function of electromagnetic field near microstrip line edge not only saves the computational resources and improves the efficiency, but also maintains the higher accuracy.
In order to solve the contradiction between the efficiency and accuracy as using the existing methods in the processing of the singularity of electromagnetic field near the microstrip line edge, in this paper, a microstrip line edge singularity processing technology based on the Locally One Dimensional Finite Difference Time Domain (LOD-FDTD), and combined with distribution function of electromagnetic field near microstrip line edge is proposed. The algorithm can handle conductors embedded in the grid area of more than 1/2 by the coordinate transformation, thus having wide applicability. Compared with the existing processing technology, the proposed algorithm in this paper has higher calculation accuracy, when the time step size is less than or equal to 5 times of the Courant-Friedrichs-Lewy (CFL) condition allowed. And compared with general LOD-FDTD, the proposed algorithm by introducing distribution function of electromagnetic field near microstrip line edge not only saves the computational resources and improves the efficiency, but also maintains the higher accuracy.
2015, 37(3): 753-757.
doi: 10.11999/JEIT140598
Abstract:
In the MIMO systems, the differential feedback is an important strategy for temporally correlated channels. A new adaptive spherical-cap differential feedback algorithm is proposed for the Single-User Multiple Input Single Output (SU-MISO) systems. Firstly, this study effectively utilizes the channels disturbance between two consecutive time instants to design the adaptive differential feedback strategy. Secondly, based on vectors scaling and composition, a new spherical-cap differential codebook algorithm is proposed, which is different from the existing methods for constructing the non-linear model. Finally, the calculation method of spherical-cap radius is deduced by combining spatial distribution of the spherical-cap codebook with channels statistical regularity. The simulation results show that the proposed algorithm can accurately track the channel disturbance and has certain performance improvement than the existing algorithms.
In the MIMO systems, the differential feedback is an important strategy for temporally correlated channels. A new adaptive spherical-cap differential feedback algorithm is proposed for the Single-User Multiple Input Single Output (SU-MISO) systems. Firstly, this study effectively utilizes the channels disturbance between two consecutive time instants to design the adaptive differential feedback strategy. Secondly, based on vectors scaling and composition, a new spherical-cap differential codebook algorithm is proposed, which is different from the existing methods for constructing the non-linear model. Finally, the calculation method of spherical-cap radius is deduced by combining spatial distribution of the spherical-cap codebook with channels statistical regularity. The simulation results show that the proposed algorithm can accurately track the channel disturbance and has certain performance improvement than the existing algorithms.
2015, 37(3): 758-762.
doi: 10.11999/JEIT140579
Abstract:
In order to broaden the antenna beamwidth, a broadband and wide beam Magneto-Electric (ME) dipole antenna with low polarization is designed based on the conventional ME antenna. By tilting and bending the dipoles, the beam width of the antenna is broadened; the consistency of the radiation pattern is improved by placing 6 parasitic patches symmetrically around the ground center. Optimizing the space between the dipoles and the length of the dipole based on the-shaped feed structure, a relative bandwidth of 58.5% (S11-10 dB) from 2.3~4.2 GHz is achieved. By optimizing the tilting angle and the parameters of the parasitic dipole, the Half Power Beam Width (HPBW) of E-plane and H-plane is simultaneously broadened to 120 in the frequency range from 2.4~4.0 GHz. The measurement results are in good agreement with the corresponding simulation, which not only prove that the broadband and wide beam characteristics of the designed antenna, but also extremely improve the consistency of the radiation pattern through the whole frequency band.
In order to broaden the antenna beamwidth, a broadband and wide beam Magneto-Electric (ME) dipole antenna with low polarization is designed based on the conventional ME antenna. By tilting and bending the dipoles, the beam width of the antenna is broadened; the consistency of the radiation pattern is improved by placing 6 parasitic patches symmetrically around the ground center. Optimizing the space between the dipoles and the length of the dipole based on the-shaped feed structure, a relative bandwidth of 58.5% (S11-10 dB) from 2.3~4.2 GHz is achieved. By optimizing the tilting angle and the parameters of the parasitic dipole, the Half Power Beam Width (HPBW) of E-plane and H-plane is simultaneously broadened to 120 in the frequency range from 2.4~4.0 GHz. The measurement results are in good agreement with the corresponding simulation, which not only prove that the broadband and wide beam characteristics of the designed antenna, but also extremely improve the consistency of the radiation pattern through the whole frequency band.
2015, 37(3): 679-685.
doi: 10.11999/JEIT140437
Abstract:
The high uncertainty and randomness are the characteristics of the sensor data in the Cyber-Physical Systems (CPS), which make the data unreliable. A creditability analysis framework is proposed to solve those problems. Abandoning the idea that the sensor is the center in modeling, the theory takes monitoring targets into consideration and constructs the sensor-target relationship diagram, which is the base of the creditability reasoning algorithm. Meanwhile, in order to reduce the space and time of searching the relationship diagram, an improving reasoning method basing on filtering the incredible targets is designed. The examples demonstrate that the proposed algorithm can filter out the false message in the sensor data and enhances the creditability of the data in CPS.
The high uncertainty and randomness are the characteristics of the sensor data in the Cyber-Physical Systems (CPS), which make the data unreliable. A creditability analysis framework is proposed to solve those problems. Abandoning the idea that the sensor is the center in modeling, the theory takes monitoring targets into consideration and constructs the sensor-target relationship diagram, which is the base of the creditability reasoning algorithm. Meanwhile, in order to reduce the space and time of searching the relationship diagram, an improving reasoning method basing on filtering the incredible targets is designed. The examples demonstrate that the proposed algorithm can filter out the false message in the sensor data and enhances the creditability of the data in CPS.
