Email alert
2014 Vol. 36, No. 8
Display Method:
2014, 36(8): 1779-1785.
doi: 10.3724/SP.J.1146.2013.01579
Abstract:
For coping with the severe inter-cell intereference in the future cellular systems, this paper studies the collaborative multicell downlink transmission for multiusers in multicell systems. This paper proposes an iterative scheme to optimize jointly relay and the Base Station (BS) beamforming weights to maximize the worst-case Signal-to-Interference-and-Noise Ratio (SINR) under total relay and BS power constraints. The proposed optimization problems for relay and BS beamforming weights can be effectively solved by using the Semi-Definite Relaxation (SDR) technology. The simulation results show that the proposed relay and BS beamforming iterative scheme can achieve near-optimal performance within a few iterations for multiusers in multicell systems.
For coping with the severe inter-cell intereference in the future cellular systems, this paper studies the collaborative multicell downlink transmission for multiusers in multicell systems. This paper proposes an iterative scheme to optimize jointly relay and the Base Station (BS) beamforming weights to maximize the worst-case Signal-to-Interference-and-Noise Ratio (SINR) under total relay and BS power constraints. The proposed optimization problems for relay and BS beamforming weights can be effectively solved by using the Semi-Definite Relaxation (SDR) technology. The simulation results show that the proposed relay and BS beamforming iterative scheme can achieve near-optimal performance within a few iterations for multiusers in multicell systems.
2014, 36(8): 1786-1791.
doi: 10.3724/SP.J.1146.2013.01371
Abstract:
For a dual-frequency Continuous Wave (CW) radar, ambiguity may occur in ranging measurement. In order to solve this problem and integrate radar and communication system, a new dual-frequency electronic system based on Extended Binary Phase Shift Keying (EBPSK)-MODEM is put forward. The system is mainly composed of two EBPSK-MODEM with different carriers, where modulator 1 transmits CW, while modulator 2 transmits EBPSK signals, but two echo signals are compared in a phase discriminator. The high-precision ranging value can be obtained. The other received EBPSK signals are also demodulated by using an impacting filer, which can convert the phase hopping into amplitude shock as the echo of pulse radar. Pulse radar echo signal can be dectected in CW background, and the output of impacting filer is wide-range measured value. After data fusion of two channel measuring, the system would output a radar measurement with both high-precision and wide-range. The radar system can also be used as a spectra efficient digital communication system by utilizing the advance of EBPSK modulator and the impacting filter aided demodulator.
For a dual-frequency Continuous Wave (CW) radar, ambiguity may occur in ranging measurement. In order to solve this problem and integrate radar and communication system, a new dual-frequency electronic system based on Extended Binary Phase Shift Keying (EBPSK)-MODEM is put forward. The system is mainly composed of two EBPSK-MODEM with different carriers, where modulator 1 transmits CW, while modulator 2 transmits EBPSK signals, but two echo signals are compared in a phase discriminator. The high-precision ranging value can be obtained. The other received EBPSK signals are also demodulated by using an impacting filer, which can convert the phase hopping into amplitude shock as the echo of pulse radar. Pulse radar echo signal can be dectected in CW background, and the output of impacting filer is wide-range measured value. After data fusion of two channel measuring, the system would output a radar measurement with both high-precision and wide-range. The radar system can also be used as a spectra efficient digital communication system by utilizing the advance of EBPSK modulator and the impacting filter aided demodulator.
2014, 36(8): 1792-1797.
doi: 10.3724/SP.J.1146.2013.01454
Abstract:
Using the second-order and third-order autocorrelation characteristics of m sequences, the maximum likehood estimation for the triple autocorrelation function of the Non-Periodic Long Code Direct Sequence Spread Spectral (NPLCDS-SS) signals is deduced. By utilizing the shift-and-add property and the triple autocorrelation characteristics of m sequences, the influence of the information code on the estimation is eliminated by the processing of delay and multiplication, a identification method for long code is proposed based on the peaks of the triple autocorrelation function in a non-cooperative communication with the certain carrier frequency and chip width of spreading code. Simulations under additive white Gaussian noise show that the correct identification probability of the algorithm is above90% when a quarter of the cycle length of the long PN code signal is used only and the signal-to-noise ratio is greater than 3.5 dB.
Using the second-order and third-order autocorrelation characteristics of m sequences, the maximum likehood estimation for the triple autocorrelation function of the Non-Periodic Long Code Direct Sequence Spread Spectral (NPLCDS-SS) signals is deduced. By utilizing the shift-and-add property and the triple autocorrelation characteristics of m sequences, the influence of the information code on the estimation is eliminated by the processing of delay and multiplication, a identification method for long code is proposed based on the peaks of the triple autocorrelation function in a non-cooperative communication with the certain carrier frequency and chip width of spreading code. Simulations under additive white Gaussian noise show that the correct identification probability of the algorithm is above90% when a quarter of the cycle length of the long PN code signal is used only and the signal-to-noise ratio is greater than 3.5 dB.
2014, 36(8): 1798-1803.
doi: 10.3724/SP.J.1146.2013.01577
Abstract:
The paper propose the definition of diffusion order and described the diffusion property of the linear permutation. Moreover, by analyzing effect of the linear permutation on the integral distinguisher length, the lower bounds for the round number of the Substitute Permutation-Generalized Feistel Structure (SP-GFS) integral distinguisher with different linear permutations P are obtained. Using this method，the integral distinguishers of two block ciphers Camellia and CLEFIA are improved, hence verifying the correctness of the conclusion.
The paper propose the definition of diffusion order and described the diffusion property of the linear permutation. Moreover, by analyzing effect of the linear permutation on the integral distinguisher length, the lower bounds for the round number of the Substitute Permutation-Generalized Feistel Structure (SP-GFS) integral distinguisher with different linear permutations P are obtained. Using this method，the integral distinguishers of two block ciphers Camellia and CLEFIA are improved, hence verifying the correctness of the conclusion.
2014, 36(8): 1804-1809.
doi: 10.3724/SP.J.1146.2014.00090
Abstract:
Quantizing the influence propagation weights between users plays an important role in commodity?marketing and promoting in social networks. However, most of current studies assume the mutual behaviors between users to influence each other are independent, while overlooked the accumulative?effect in influence propagation process. To fill this gap, this study proposes an influence weights learning approach under the framework of the linear threshold model. With a log of past propagations of involved users in social networks, the study formulize an objective function on the basis of maximum likelihood estimation for the proposed problem, and presents a particle swarm optimization algorithm according to the objective function. Experimental results on real-world datasets validate the effectiveness of the proposed approach.
Quantizing the influence propagation weights between users plays an important role in commodity?marketing and promoting in social networks. However, most of current studies assume the mutual behaviors between users to influence each other are independent, while overlooked the accumulative?effect in influence propagation process. To fill this gap, this study proposes an influence weights learning approach under the framework of the linear threshold model. With a log of past propagations of involved users in social networks, the study formulize an objective function on the basis of maximum likelihood estimation for the proposed problem, and presents a particle swarm optimization algorithm according to the objective function. Experimental results on real-world datasets validate the effectiveness of the proposed approach.
2014, 36(8): 1810-1816.
doi: 10.3724/SP.J.1146.2013.01362
Abstract:
The most recent studies in the trust networks focus on the trust inference and aggregation mechanisms, but the issues of correlations between trusted nodes and their structural analysis have not get much attention. To address this weakness, a new Multi-dimensional Trust Sequential Pattern (MTSP) mining algorithm called is proposed, which mainly includes two continuous processes: mining the frequent trust sequences and then filtering the multi-dimensional patterns. And with multiple factors such as trust strength, length of sequences and node credibility taken into account, the algorithm can effectively grab the multi-dimensional frequent trust sequences in the trust networks that imply the correlations between the important nodes as well as their sequence structure in these trust sequences. The simulation experiments show that the results of the proposed MTSP algorithm is able to comprehensively and accurately reflect the characteristics of the important nodes and correlations between them in the trust networks.
The most recent studies in the trust networks focus on the trust inference and aggregation mechanisms, but the issues of correlations between trusted nodes and their structural analysis have not get much attention. To address this weakness, a new Multi-dimensional Trust Sequential Pattern (MTSP) mining algorithm called is proposed, which mainly includes two continuous processes: mining the frequent trust sequences and then filtering the multi-dimensional patterns. And with multiple factors such as trust strength, length of sequences and node credibility taken into account, the algorithm can effectively grab the multi-dimensional frequent trust sequences in the trust networks that imply the correlations between the important nodes as well as their sequence structure in these trust sequences. The simulation experiments show that the results of the proposed MTSP algorithm is able to comprehensively and accurately reflect the characteristics of the important nodes and correlations between them in the trust networks.
2014, 36(8): 1817-1823.
doi: 10.3724/SP.J.1146.2013.01417
Abstract:
In order to improve the response performance and decrease the transmission interruption probability, a centralized spectrum allocation strategy with an access threshold and super-slot for cognitive radio networks is proposed by limiting the number of data packets in the system and allowing the primary users access spectrum with a pre-emptive priority based on a super-slot. Through combining the sequence number of the specified slot in a super-slot and the number of data packets for the cognitive users in the system, a two-dimensional discrete-time Markov chain model is established presenting the formulas for the performance measures in terms of the average latency, the throughput and the channel switching ratio as well. A profit function is created; the access threshold is optimized for different super-slot sizes and the system experiments are conducted via the use of the optimization results. The experiment results show that a reasonable access threshold and a proper super-slot size can improve the transmission quality for the cognitive users.
In order to improve the response performance and decrease the transmission interruption probability, a centralized spectrum allocation strategy with an access threshold and super-slot for cognitive radio networks is proposed by limiting the number of data packets in the system and allowing the primary users access spectrum with a pre-emptive priority based on a super-slot. Through combining the sequence number of the specified slot in a super-slot and the number of data packets for the cognitive users in the system, a two-dimensional discrete-time Markov chain model is established presenting the formulas for the performance measures in terms of the average latency, the throughput and the channel switching ratio as well. A profit function is created; the access threshold is optimized for different super-slot sizes and the system experiments are conducted via the use of the optimization results. The experiment results show that a reasonable access threshold and a proper super-slot size can improve the transmission quality for the cognitive users.
2014, 36(8): 1824-1830.
doi: 10.3724/SP.J.1146.2013.01398
Abstract:
6LoWPAN protocol supports IPv6 datagram delivering over IEEE 802.15.4-based low-power wireless personal area network, and its main functions include fragmenting and reassembling IPv6 datagrams, header compression, and routing. In this paper, the number of transmissions, the delay and the throughput of multi-path routing for delivering an IPv6 datagram to the destination from the source are derived based on probability theory. Moreover, the model that optimizes the throughput of multi-path routing is developed, and based on this model the optimal fragmentation scheme is proposed. The proposed scheme is able to improve the throughput of IPv6 datagrams in 6LoWPAN.
6LoWPAN protocol supports IPv6 datagram delivering over IEEE 802.15.4-based low-power wireless personal area network, and its main functions include fragmenting and reassembling IPv6 datagrams, header compression, and routing. In this paper, the number of transmissions, the delay and the throughput of multi-path routing for delivering an IPv6 datagram to the destination from the source are derived based on probability theory. Moreover, the model that optimizes the throughput of multi-path routing is developed, and based on this model the optimal fragmentation scheme is proposed. The proposed scheme is able to improve the throughput of IPv6 datagrams in 6LoWPAN.
2014, 36(8): 1831-1837.
doi: 10.3724/SP.J.1146.2013.01272
Abstract:
A multi-dimension association analysis method of users behavioral characteristics based on association rules is proposed for the discovery of information content security incidents in network. The users multi- dimension data which generate in communication can be mined. An inspection standard based on Bonferronis correction is put forward to deal with the problem of false alarm. In order to meet the demand for the implementation of the method in a massive database, a distributed power set Apriori algorithm in Map-Reduce framework is proposed. Experimental results demonstrate that the proposed method and its corresponding algorithm have strong ability in parallel computing. The algorithm has a great detection rate in the case of low false alarm rate and missing detection rate. The running time is short and it can achieve a fast convergences rate.
A multi-dimension association analysis method of users behavioral characteristics based on association rules is proposed for the discovery of information content security incidents in network. The users multi- dimension data which generate in communication can be mined. An inspection standard based on Bonferronis correction is put forward to deal with the problem of false alarm. In order to meet the demand for the implementation of the method in a massive database, a distributed power set Apriori algorithm in Map-Reduce framework is proposed. Experimental results demonstrate that the proposed method and its corresponding algorithm have strong ability in parallel computing. The algorithm has a great detection rate in the case of low false alarm rate and missing detection rate. The running time is short and it can achieve a fast convergences rate.
2014, 36(8): 1838-1843.
doi: 10.3724/SP.J.1146.2013.01416
Abstract:
Sparse representation based visual trackers are very computationally inefficient and prone to model drifting. To deal with these issues, a novel visual tracking method is proposed based on L2 -norm regularized robust coding. The proposed method solves the coding coefficient of candidate objects via robust coding based on L2-norm regularization, and it achieves visual tracking by taking weighted reconstruction errors of the candidate object as observation likelihood in particle filter framework. In addition, to adapt the changes of object appearance and avoid model drifting, an occlusion detection method for template update is proposed by investigating the weight matrix of current object estimated with L2-norm regularized robust coding. The experimental results on several challenging sequences show that the proposed method has better performance than that of the state-of-the-art tracker.
Sparse representation based visual trackers are very computationally inefficient and prone to model drifting. To deal with these issues, a novel visual tracking method is proposed based on L2 -norm regularized robust coding. The proposed method solves the coding coefficient of candidate objects via robust coding based on L2-norm regularization, and it achieves visual tracking by taking weighted reconstruction errors of the candidate object as observation likelihood in particle filter framework. In addition, to adapt the changes of object appearance and avoid model drifting, an occlusion detection method for template update is proposed by investigating the weight matrix of current object estimated with L2-norm regularized robust coding. The experimental results on several challenging sequences show that the proposed method has better performance than that of the state-of-the-art tracker.
2014, 36(8): 1844-1851.
doi: 10.3724/SP.J.1146.2013.01389
Abstract:
Person re-identification is among the key issues in video surveillance. From the viewpoint of fusing appearance statistical features, human color and structure information are exploited; two statistical descriptors named spatiogram and region covariance are both explored on feature designing and metric choosing. Several complimentary feature vectors are extracted from a proper number of hierarchical image layers and regions. The simplest l1 norm distance is chosen to form the proposed weighted combining distance. The fused method with such two descriptors requires neither preprocessing nor supervised training. Extensive experiments by comparisons and analysis show that the proposed method not only achieves the state-of-the-art re-identification performance, but also enjoys a great applicability.
Person re-identification is among the key issues in video surveillance. From the viewpoint of fusing appearance statistical features, human color and structure information are exploited; two statistical descriptors named spatiogram and region covariance are both explored on feature designing and metric choosing. Several complimentary feature vectors are extracted from a proper number of hierarchical image layers and regions. The simplest l1 norm distance is chosen to form the proposed weighted combining distance. The fused method with such two descriptors requires neither preprocessing nor supervised training. Extensive experiments by comparisons and analysis show that the proposed method not only achieves the state-of-the-art re-identification performance, but also enjoys a great applicability.
2014, 36(8): 1852-1858.
doi: 10.3724/SP.J.1146.2013.01614
Abstract:
As to the difficulty of confidence measure estimation regarding to Automatic Speech Recognition (ASR), a strategy resorting to multi-source knowledge combination to improve the confidence measure is proposed in this paper. More specially, the knowledge come from acoustic level, linguistic level and semantic level are firstly selected and then combined by different ways by held-out validation. And then, these multi-source knowledge are integrated under the framework of Hidden-units Conditional Random Fields (HuCRFs). Lastly, the conditional probability computed from HuCRFs is used to be a new estimation procedure of confidence measure for recognition candidate. Experiments show that the discriminative ability of conditional probability of HuCRFs is superior to the conventional posterior computed from lattice. Furthermore, a lattice rescoring is carried out by utilizing the conditional probabilities of HuCRFs to search the best hypotheses and resulted in a significant reduction on Character Error Rate (CER) by about 2% absolutely on a benchmark corpus. Simultaneously, a performance comparison between the performances of long-distance language model based lattice rescoring and conditional probability of HuCRFs based lattice rescoring is also performed and it is further proved that HuCRFs is a better alternative to the estimation of confidence measure in ASR.
As to the difficulty of confidence measure estimation regarding to Automatic Speech Recognition (ASR), a strategy resorting to multi-source knowledge combination to improve the confidence measure is proposed in this paper. More specially, the knowledge come from acoustic level, linguistic level and semantic level are firstly selected and then combined by different ways by held-out validation. And then, these multi-source knowledge are integrated under the framework of Hidden-units Conditional Random Fields (HuCRFs). Lastly, the conditional probability computed from HuCRFs is used to be a new estimation procedure of confidence measure for recognition candidate. Experiments show that the discriminative ability of conditional probability of HuCRFs is superior to the conventional posterior computed from lattice. Furthermore, a lattice rescoring is carried out by utilizing the conditional probabilities of HuCRFs to search the best hypotheses and resulted in a significant reduction on Character Error Rate (CER) by about 2% absolutely on a benchmark corpus. Simultaneously, a performance comparison between the performances of long-distance language model based lattice rescoring and conditional probability of HuCRFs based lattice rescoring is also performed and it is further proved that HuCRFs is a better alternative to the estimation of confidence measure in ASR.
2014, 36(8): 1859-1865.
doi: 10.3724/SP.J.1146.2013.01468
Abstract:
Three-dimensional Minimum Error Thresholding (3D-MET) is more robust to noise than MET and 2D-MET, but its computational complexity grows exponentially. By constructing look-up tables recursively, its fast algorithm 3D-RMET reduces the complexity from O(L6) to O(L3), but its complexity is still too high to be applied to the project. A novel fast method is proposed based on dimension reduction and grading strategy. Firstly, based on the decomposition of 3D-MET, a new threshold discriminant is proposed to reduce the dimensionality from 3D to 1D. And then, the 3D histogram of test image is grouped and rebuilt to further improve its processing speed. Finally, segmentation results of 3D-MET, 3D-RMET and the proposed method are given and evaluated by performance criteria. Experiments and evaluation results indicate that without losing the robustness to noise, the proposed method reduces the time complexity from O(L6) to O(L1/2). Compared with 3D-RMET, the proposed method is 6 magnitudes faster than the former.
Three-dimensional Minimum Error Thresholding (3D-MET) is more robust to noise than MET and 2D-MET, but its computational complexity grows exponentially. By constructing look-up tables recursively, its fast algorithm 3D-RMET reduces the complexity from O(L6) to O(L3), but its complexity is still too high to be applied to the project. A novel fast method is proposed based on dimension reduction and grading strategy. Firstly, based on the decomposition of 3D-MET, a new threshold discriminant is proposed to reduce the dimensionality from 3D to 1D. And then, the 3D histogram of test image is grouped and rebuilt to further improve its processing speed. Finally, segmentation results of 3D-MET, 3D-RMET and the proposed method are given and evaluated by performance criteria. Experiments and evaluation results indicate that without losing the robustness to noise, the proposed method reduces the time complexity from O(L6) to O(L1/2). Compared with 3D-RMET, the proposed method is 6 magnitudes faster than the former.
2014, 36(8): 1866-1871.
doi: 10.3724/SP.J.1146.2014.00154
Abstract:
This study firstly analyzes the model for Poisson noise removal by Le et al. (2007) from the view of calculus of variations, and gets a box constraint of the solution to the model. Then by incorporating the Alternating Direction Method of Multipliers (ADMM) algorithm, a fast total variation algorithm based on box constraint is proposed to solve the above model numerically, and the convergence of the fast algorithm is proved. Finally, experimental results are reported to demonstrate the feasibility and effectiveness of this algorithm.
This study firstly analyzes the model for Poisson noise removal by Le et al. (2007) from the view of calculus of variations, and gets a box constraint of the solution to the model. Then by incorporating the Alternating Direction Method of Multipliers (ADMM) algorithm, a fast total variation algorithm based on box constraint is proposed to solve the above model numerically, and the convergence of the fast algorithm is proved. Finally, experimental results are reported to demonstrate the feasibility and effectiveness of this algorithm.
2014, 36(8): 1872-1877.
doi: 10.3724/SP.J.1146.2013.01459
Abstract:
The traditional algorithm of the Per-Survivor Processing (PSP) model is analyzed, and the improved algorithm of nonbinary Soft-Output Viterbi Algorithm and PSP (SOVA-PSP) with soft output for feedforward are given. For the condition of the single way timing accurate, the SOVA-PSP algorithm on blind separation of Paired Carrier Multiple Access (PCMA) signals is presented. Relative to the traditional PSP algorithm, the state of the proposed algorithm is reduced from M2(L-1) to M(L-1) (M is the order of the modulator, L is the length of equivalent channel response), so as to greatly reduce the algorithm complexity. Simulation results demonstrate that, compared with the traditional algorithm of PSP, the improved algorithm can reduce the complexity with almost no performance loss.
The traditional algorithm of the Per-Survivor Processing (PSP) model is analyzed, and the improved algorithm of nonbinary Soft-Output Viterbi Algorithm and PSP (SOVA-PSP) with soft output for feedforward are given. For the condition of the single way timing accurate, the SOVA-PSP algorithm on blind separation of Paired Carrier Multiple Access (PCMA) signals is presented. Relative to the traditional PSP algorithm, the state of the proposed algorithm is reduced from M2(L-1) to M(L-1) (M is the order of the modulator, L is the length of equivalent channel response), so as to greatly reduce the algorithm complexity. Simulation results demonstrate that, compared with the traditional algorithm of PSP, the improved algorithm can reduce the complexity with almost no performance loss.
2014, 36(8): 1878-1883.
doi: 10.3724/SP.J.1146.2013.01436
Abstract:
Conventional time-frequency analysis is a powerful tool for Frequency-Hopping (FH) signal processing, however, it fails to realize the parameter estimation in stable noise environment. Time-frequency analysis based on Merid filter is proposed for FH signal parameter estimation. Merid filter can suppress effectively the stable noise. The observed signal is processed by Merid filter at first, then the FH signal parameters are estimated by Short-Time Fourier Transform (STFT). Simulation results show that the proposed method has better parameter estimation performance for FH signals than the fractional lower order statistics as well as the Myriad filter based time frequency analysis methods in stable noise environment.
Conventional time-frequency analysis is a powerful tool for Frequency-Hopping (FH) signal processing, however, it fails to realize the parameter estimation in stable noise environment. Time-frequency analysis based on Merid filter is proposed for FH signal parameter estimation. Merid filter can suppress effectively the stable noise. The observed signal is processed by Merid filter at first, then the FH signal parameters are estimated by Short-Time Fourier Transform (STFT). Simulation results show that the proposed method has better parameter estimation performance for FH signals than the fractional lower order statistics as well as the Myriad filter based time frequency analysis methods in stable noise environment.
2014, 36(8): 1884-1890.
doi: 10.3724/SP.J.1146.2013.01446
Abstract:
A Guided Self-adaptive Evolutionary Genetic Algorithm (GSEGA) is proposed. The principle of good point set is used to generate the initial population. Based on the elitist preserved method, a way of parallel crossing and mutation with population-segmentation is offered, in which a son population among the segmented population is randomly generated. In addition, a guided self-adaptive mutation strategy based on the statistics of the more excellent individualities is adopted on the other part of the son population to speed up the evolution. Through the use of the homogeneous finite Markov chain model, the global convergence and high searching speed of the GSEGA is proved. The experimental results show that the GSEGA presents a higher speed and precision in comparison with the other Genetic Algorithms (GAs).
A Guided Self-adaptive Evolutionary Genetic Algorithm (GSEGA) is proposed. The principle of good point set is used to generate the initial population. Based on the elitist preserved method, a way of parallel crossing and mutation with population-segmentation is offered, in which a son population among the segmented population is randomly generated. In addition, a guided self-adaptive mutation strategy based on the statistics of the more excellent individualities is adopted on the other part of the son population to speed up the evolution. Through the use of the homogeneous finite Markov chain model, the global convergence and high searching speed of the GSEGA is proved. The experimental results show that the GSEGA presents a higher speed and precision in comparison with the other Genetic Algorithms (GAs).
2014, 36(8): 1891-1898.
doi: 10.3724/SP.J.1146.2013.01433
Abstract:
In this paper, a novel framework for remote sensing image annotation is proposed based on spatial constrained multi-feature joint sparse coding to extend the sparse representation-based classifier to multi-feature framework. The proposed framework imposed an l1,2 mixed-norm regularization on encode coefficients of multiple features. The regularization encourages the coefficients to share a common sparsity pattern, which preserves the cross-feature information. Inspired by the success of dictionary learning, a novel dictionary learning model is proposed to promote the performance of multi-feature joint sparse coding, while the cross-feature association is preserved by consistent transformation constraint. In addition, spatial dependencies between patches of remote sensing images are useful for annotation task but usually ignored of insufficiently exploited. In this paper, a spatial relation constrained classifier is designed to incorporate spatial coherence into multi-feature sparse coding model to annotate images more precisely. Experiments on public dataset and large satellite images show the discriminative power and effectiveness of the proposed framework.
In this paper, a novel framework for remote sensing image annotation is proposed based on spatial constrained multi-feature joint sparse coding to extend the sparse representation-based classifier to multi-feature framework. The proposed framework imposed an l1,2 mixed-norm regularization on encode coefficients of multiple features. The regularization encourages the coefficients to share a common sparsity pattern, which preserves the cross-feature information. Inspired by the success of dictionary learning, a novel dictionary learning model is proposed to promote the performance of multi-feature joint sparse coding, while the cross-feature association is preserved by consistent transformation constraint. In addition, spatial dependencies between patches of remote sensing images are useful for annotation task but usually ignored of insufficiently exploited. In this paper, a spatial relation constrained classifier is designed to incorporate spatial coherence into multi-feature sparse coding model to annotate images more precisely. Experiments on public dataset and large satellite images show the discriminative power and effectiveness of the proposed framework.
2014, 36(8): 1899-1904.
doi: 10.3724/SP.J.1146.2013.01588
Abstract:
A Range Doppler imaging Algorithm (RDA) based on Graphic Processing Unit (GPU) is proposed, which is a new method of real-time imaging for Synthetic Aperture Sonar (SAS). The range compression, phase compensation and azimuth compression are implemented in parallel based on GPU device, which increases greatly the efficiency of Range Doppler Algorithm (RDA). The simulation and experimental results show that the GPU- based implementation is faster (up to 22 times) than the CPU-based implementation, which makes it suitable for real-time SAS imaging.
A Range Doppler imaging Algorithm (RDA) based on Graphic Processing Unit (GPU) is proposed, which is a new method of real-time imaging for Synthetic Aperture Sonar (SAS). The range compression, phase compensation and azimuth compression are implemented in parallel based on GPU device, which increases greatly the efficiency of Range Doppler Algorithm (RDA). The simulation and experimental results show that the GPU- based implementation is faster (up to 22 times) than the CPU-based implementation, which makes it suitable for real-time SAS imaging.
2014, 36(8): 1905-1911.
doi: 10.3724/SP.J.1146.2013.01456
Abstract:
A novel parametric focusing algorithm, using the Cross Spectrum MUSIC (CSMUSIC) method, is presented to obtain a finely focused image of ground moving target on spaceborne distributed SAR. By exploiting the space-time property of the azimuth echo signal, an Extended Space-Time Model (ESTM) is firstly introduced. Then on the basis of this model, a parametric estimating method based on subspace theory is presented to estimate the azimuth velocity of moving target and to obtain an image of finely focused. The proposed method has higher precision and lower calculation burden than the conventional algorithms. The simulation results indicate that the proposed method is verified to be more efficient in contrast with the conventional algorithms.
A novel parametric focusing algorithm, using the Cross Spectrum MUSIC (CSMUSIC) method, is presented to obtain a finely focused image of ground moving target on spaceborne distributed SAR. By exploiting the space-time property of the azimuth echo signal, an Extended Space-Time Model (ESTM) is firstly introduced. Then on the basis of this model, a parametric estimating method based on subspace theory is presented to estimate the azimuth velocity of moving target and to obtain an image of finely focused. The proposed method has higher precision and lower calculation burden than the conventional algorithms. The simulation results indicate that the proposed method is verified to be more efficient in contrast with the conventional algorithms.
2014, 36(8): 1912-1918.
doi: 10.3724/SP.J.1146.2013.01744
Abstract:
Firstly, in the framework of the Interacting Multiple Model (IMM) algorithm the constant-velocity model and the adaptive constant acceleration model are selected as the dynamic models for the un-maneuvering and maneuvering states of the un-cooperative target, which is called the adaptive IMM algorithm. Then since it is necessary to consider the performance of estimating range/velocity and Doppler tolerance for tracking a maneuvering target, the V-Linear Frequency Modulated (V-LFM) signal is selected as the transmitted signal in the radar system. The analysis on the Cramer-Rao Lower Bound (CRLB) for estimating the range/velocity and Doppler tolerance of three signals (LFM, V-LFM and M sequence) shows that the V-LFM waveform can effectively improve the performance of estimating the target range and velocity in the case of a bit loss in the Doppler tolerance. The simulations demonstrate that the tracking performance is apparently improved, when multiple pulses of V-LFM waveform is transmitted and the adaptive IMM algorithm is utilized in the radar system.
Firstly, in the framework of the Interacting Multiple Model (IMM) algorithm the constant-velocity model and the adaptive constant acceleration model are selected as the dynamic models for the un-maneuvering and maneuvering states of the un-cooperative target, which is called the adaptive IMM algorithm. Then since it is necessary to consider the performance of estimating range/velocity and Doppler tolerance for tracking a maneuvering target, the V-Linear Frequency Modulated (V-LFM) signal is selected as the transmitted signal in the radar system. The analysis on the Cramer-Rao Lower Bound (CRLB) for estimating the range/velocity and Doppler tolerance of three signals (LFM, V-LFM and M sequence) shows that the V-LFM waveform can effectively improve the performance of estimating the target range and velocity in the case of a bit loss in the Doppler tolerance. The simulations demonstrate that the tracking performance is apparently improved, when multiple pulses of V-LFM waveform is transmitted and the adaptive IMM algorithm is utilized in the radar system.
2014, 36(8): 1919-1925.
doi: 10.3724/SP.J.1146.2013.01450
Abstract:
Rotational motion estimation is essential for MIMO radar imaging with ISAR technique. According to the estimation, the echo data can be rearranged and interpolated, and the cross-range scaling can be implemented for range-Doppler imaging. For an object rotating with a constant acceleration, a method is proposed to jointly estimate the initial rotating velocity and the rotating acceleration. It estimates the phase factors of the difference signal by exploiting the phase difference between the echo signals from two different channels of MIMO radar. Based on this, the errors induced by trigonometric approximation in the derivation of the method are analyzed, and then the influencing factors causing these errors are obtained. Meanwhile, the motion parameter estimation resolutions are assessed quantitatively. Finally, simulations are performed to verify the correctness of the proposed method and the analysis.
Rotational motion estimation is essential for MIMO radar imaging with ISAR technique. According to the estimation, the echo data can be rearranged and interpolated, and the cross-range scaling can be implemented for range-Doppler imaging. For an object rotating with a constant acceleration, a method is proposed to jointly estimate the initial rotating velocity and the rotating acceleration. It estimates the phase factors of the difference signal by exploiting the phase difference between the echo signals from two different channels of MIMO radar. Based on this, the errors induced by trigonometric approximation in the derivation of the method are analyzed, and then the influencing factors causing these errors are obtained. Meanwhile, the motion parameter estimation resolutions are assessed quantitatively. Finally, simulations are performed to verify the correctness of the proposed method and the analysis.
2014, 36(8): 1926-1931.
doi: 10.3724/SP.J.1146.2013.01560
Abstract:
A hybrid signal processing architecture characterized by all transmit and any receive is proposed firstly for distributed aperture coherent radar when receivers are co-located with partial transmitters. Then the closed-form Cramer-Rao Bounds (CRB) of coherence parameters estimation are derived under the condition that multiple pulses are transmitted, and the relationship among estimation performance and the number of antennas and pulses is studied. It is concluded that the CRB of parameters estimation for co-located antennas is slower than that for bistatic antennas, and the CRB reduces with the increase of the number of antennas and pulses. Finally, the numerical examples demonstrate the validity of the theoretical results.
A hybrid signal processing architecture characterized by all transmit and any receive is proposed firstly for distributed aperture coherent radar when receivers are co-located with partial transmitters. Then the closed-form Cramer-Rao Bounds (CRB) of coherence parameters estimation are derived under the condition that multiple pulses are transmitted, and the relationship among estimation performance and the number of antennas and pulses is studied. It is concluded that the CRB of parameters estimation for co-located antennas is slower than that for bistatic antennas, and the CRB reduces with the increase of the number of antennas and pulses. Finally, the numerical examples demonstrate the validity of the theoretical results.
2014, 36(8): 1932-1938.
doi: 10.3724/SP.J.1146.2013.01231
Abstract:
A phase-based method using a single channel SAR is proposed to estimate the radial velocity of ground moving targets unambiguously. Two-look operation in the range frequency domain and interferometry in the Doppler domain are done made to keep the phase meet the assumption of phase continuity. The lease squares linear fitting is used to estimate the slope between interferometric phase and Doppler frequency, and the radial velocity can be calculated by the slope. The proposed method possesses the advantage of being independent on phase wrapping. Compared with the single channel amplitude-based method, the method is able to provide greater precision and usefulness. The radial velocities of multi-targets can be estimated simultaneously in the Doppler domain. Experimental results validate the effectiveness of the proposed method.
A phase-based method using a single channel SAR is proposed to estimate the radial velocity of ground moving targets unambiguously. Two-look operation in the range frequency domain and interferometry in the Doppler domain are done made to keep the phase meet the assumption of phase continuity. The lease squares linear fitting is used to estimate the slope between interferometric phase and Doppler frequency, and the radial velocity can be calculated by the slope. The proposed method possesses the advantage of being independent on phase wrapping. Compared with the single channel amplitude-based method, the method is able to provide greater precision and usefulness. The radial velocities of multi-targets can be estimated simultaneously in the Doppler domain. Experimental results validate the effectiveness of the proposed method.
2014, 36(8): 1939-1945.
doi: 10.3724/SP.J.1146.2013.01455
Abstract:
Track association is a precondition of the distributed multi-sensors track fusion. Given the fact that the fusion center is not able to get the target states estimation covariance, a global optimal track association algorithm based on sequential modified grey association degree is proposed. The algorithm cancels the scope normalization, sequentially accumulates data array index absolute difference and modifies the grey association coefficient formulation to ensure exchangeability, thus yielding the sequential modified grey association degree between the sensors tracks. Then the global optimal track association is obtained by making the association degree as the global statistical vector. The simulation results show that the performance and robustness of the proposed algorithm is apparently better than the traditional algorithm under the condition of dense parallel formation, random cross targets and unshared observation in existence.
Track association is a precondition of the distributed multi-sensors track fusion. Given the fact that the fusion center is not able to get the target states estimation covariance, a global optimal track association algorithm based on sequential modified grey association degree is proposed. The algorithm cancels the scope normalization, sequentially accumulates data array index absolute difference and modifies the grey association coefficient formulation to ensure exchangeability, thus yielding the sequential modified grey association degree between the sensors tracks. Then the global optimal track association is obtained by making the association degree as the global statistical vector. The simulation results show that the performance and robustness of the proposed algorithm is apparently better than the traditional algorithm under the condition of dense parallel formation, random cross targets and unshared observation in existence.
2014, 36(8): 1946-1953.
doi: 10.3724/SP.J.1146.2013.01415
Abstract:
Ultra-WideBand Multi-Input-Multi-Output (UWB MIMO) radar plays more and more important role in through-the-wall application presently, because of its well resolution performance, limited elements requirements and efficient ability for data capturing. However, since its characteristics and increased requirements, it put forward new demands for the corresponding channel construction. This study reconstruct the channel for UWB MIMO through-the-wall radar, following analyzed its characteristics. This study extend the scene form the conventional single-wall-based to building-based. Then this study compute the signal distance of each echo component and derive echo expression based on extended theory. Based on the proposed model, an efficient moving target imaging method is proposed in this paper. Using the TWI equipment in actual environment, the accuracy of the model is validated and the efficient performance in improving image quality is also demonstrated.
Ultra-WideBand Multi-Input-Multi-Output (UWB MIMO) radar plays more and more important role in through-the-wall application presently, because of its well resolution performance, limited elements requirements and efficient ability for data capturing. However, since its characteristics and increased requirements, it put forward new demands for the corresponding channel construction. This study reconstruct the channel for UWB MIMO through-the-wall radar, following analyzed its characteristics. This study extend the scene form the conventional single-wall-based to building-based. Then this study compute the signal distance of each echo component and derive echo expression based on extended theory. Based on the proposed model, an efficient moving target imaging method is proposed in this paper. Using the TWI equipment in actual environment, the accuracy of the model is validated and the efficient performance in improving image quality is also demonstrated.
2014, 36(8): 1954-1959.
doi: 10.3724/SP.J.1146.2013.01281
Abstract:
A reduced-dimensional Direction Of Arrival (DOA) estimation method in monostatic MIMO radar with L-shaped array based on MUSIC algorithm is proposed. Firstly, considering of the steering vector of L-shaped array, a reduced-dimensional matrix is employed to transform the data matrix into a low dimensional space. Then the quadratic optimization algorithm is utilized to decompose two-dimensional (2-D) DOA estimation into two one-dimensional (1-D) DOA estimation. Finally, one of the two angles is estimated by MUSIC spatial spectrum, then the other angle is estimated by Root-MUSIC method with the spatial spectrum function. The proposed method which only requires 1-D search, can greatly avoid the high computational cost within 2-D MUSIC algorithm. Simulation results verify its correctness and feasibility.
A reduced-dimensional Direction Of Arrival (DOA) estimation method in monostatic MIMO radar with L-shaped array based on MUSIC algorithm is proposed. Firstly, considering of the steering vector of L-shaped array, a reduced-dimensional matrix is employed to transform the data matrix into a low dimensional space. Then the quadratic optimization algorithm is utilized to decompose two-dimensional (2-D) DOA estimation into two one-dimensional (1-D) DOA estimation. Finally, one of the two angles is estimated by MUSIC spatial spectrum, then the other angle is estimated by Root-MUSIC method with the spatial spectrum function. The proposed method which only requires 1-D search, can greatly avoid the high computational cost within 2-D MUSIC algorithm. Simulation results verify its correctness and feasibility.
2014, 36(8): 1960-1965.
doi: 10.3724/SP.J.1146.2013.01478
Abstract:
Due to the high orbit and long synthetic aperture time of Geosynchronous SAR (Geo-SAR), the error brought by the Stop-Go assumption has to be considered. In this paper, an accurate signal propagation delay time equation of Geo-SAR based on circle orbit is presented, and the error of Stop-Go assumption is shown by simulation. An equivalent position model is proposed, and is validated through simulation. Error of different powers of Stop-Go slant range is analyzed using Legendre orthogonal series expansion. The limitation of Stop-Go assumption and the availability of Equivalent Position mode are proved by point target simulation.
Due to the high orbit and long synthetic aperture time of Geosynchronous SAR (Geo-SAR), the error brought by the Stop-Go assumption has to be considered. In this paper, an accurate signal propagation delay time equation of Geo-SAR based on circle orbit is presented, and the error of Stop-Go assumption is shown by simulation. An equivalent position model is proposed, and is validated through simulation. Error of different powers of Stop-Go slant range is analyzed using Legendre orthogonal series expansion. The limitation of Stop-Go assumption and the availability of Equivalent Position mode are proved by point target simulation.
2014, 36(8): 1966-1971.
doi: 10.3724/SP.J.1146.2013.01524
Abstract:
For the high sidelobe of waveforms after pulse compression of the numerical optimization methods in the designing orthogonal waveform for MIMO radar, a method of Golay complementary sequence with space time coding is proposed. The autocorrelation sidelobe and crosscorrelation of the waveform are degraded by utilizing the complementary characteristic of the Golay complementary sequences and extended space time coding the pulse trains of the sequences, as a result, the sidelobe level of the waveform after pulse compression is greatly reduced. Then, to solve the problem of the decrease of target detection ability for Doppler shift, the waveform after pulse compression is Doppler compensated via null space vector weighting method, consequently, the target detection ability of the waveform is significantly improved. The simulation results demonstrate the effectiveness of the method.
For the high sidelobe of waveforms after pulse compression of the numerical optimization methods in the designing orthogonal waveform for MIMO radar, a method of Golay complementary sequence with space time coding is proposed. The autocorrelation sidelobe and crosscorrelation of the waveform are degraded by utilizing the complementary characteristic of the Golay complementary sequences and extended space time coding the pulse trains of the sequences, as a result, the sidelobe level of the waveform after pulse compression is greatly reduced. Then, to solve the problem of the decrease of target detection ability for Doppler shift, the waveform after pulse compression is Doppler compensated via null space vector weighting method, consequently, the target detection ability of the waveform is significantly improved. The simulation results demonstrate the effectiveness of the method.
2014, 36(8): 1972-1977.
doi: 10.3724/SP.J.1146.2013.01573
Abstract:
The paper proposes a method of two-dimensional attitude steering to compensate Doppler centroid in Geosynchronous Earth Orbit Synthetic Aperture Radar (GEO SAR). By guaranteeing the beam illuminating the earths surface, Doppler centroid is compensated to zero by controlling the beam center direction. Firstly, for showing the necessity of two-dimensional attitude steering, Doppler centroid without yaw control and under one-dimensional yaw steering are analyzed. Then, based on the intersection line between the zero-Doppler plane and the earths surface, the calculation expression of beam center direction at any position of the satellite is deduced to compensate Doppler centroid to zero under the satellite obit coordinate. Simulation experiments validate the effectiveness of the proposed method.
The paper proposes a method of two-dimensional attitude steering to compensate Doppler centroid in Geosynchronous Earth Orbit Synthetic Aperture Radar (GEO SAR). By guaranteeing the beam illuminating the earths surface, Doppler centroid is compensated to zero by controlling the beam center direction. Firstly, for showing the necessity of two-dimensional attitude steering, Doppler centroid without yaw control and under one-dimensional yaw steering are analyzed. Then, based on the intersection line between the zero-Doppler plane and the earths surface, the calculation expression of beam center direction at any position of the satellite is deduced to compensate Doppler centroid to zero under the satellite obit coordinate. Simulation experiments validate the effectiveness of the proposed method.
2014, 36(8): 1978-1984.
doi: 10.3724/SP.J.1146.2013.01936
Abstract:
Active radar can improve the cross-range resolution in the squint and side-looking direction using Doppler Beam Sharpening (DBS) and Synthetic Aperture Radar (SAR) techniques, but can difficultly determine multiple targets in the forward-looking direction at the same range cell but different angel in one beam. A spare target-scenario determination strategy based on Compressive Sensing (CS) framework is addressed which can obtain the resolution in the line of sight for active radar and need only one receiver channel. The outputs of multiple sub-arrays are randomly weighted and then summarized in single receiver channel. The single-receiver outputs in the same one range cell belonging to multiple pulse repetition periods are modeled as multiple observations with respect to the same one spare target-scenario, and then the sparse-target scene is estimated using the recovery method of compressive sensing based on observations. Numerical experiments show that the proposed method can obtain the resolution of a spare target scenario in one beam in the line of sight for active radar.
Active radar can improve the cross-range resolution in the squint and side-looking direction using Doppler Beam Sharpening (DBS) and Synthetic Aperture Radar (SAR) techniques, but can difficultly determine multiple targets in the forward-looking direction at the same range cell but different angel in one beam. A spare target-scenario determination strategy based on Compressive Sensing (CS) framework is addressed which can obtain the resolution in the line of sight for active radar and need only one receiver channel. The outputs of multiple sub-arrays are randomly weighted and then summarized in single receiver channel. The single-receiver outputs in the same one range cell belonging to multiple pulse repetition periods are modeled as multiple observations with respect to the same one spare target-scenario, and then the sparse-target scene is estimated using the recovery method of compressive sensing based on observations. Numerical experiments show that the proposed method can obtain the resolution of a spare target scenario in one beam in the line of sight for active radar.
2014, 36(8): 1985-1991.
doi: 10.3724/SP.J.1146.2013.01466
Abstract:
As it is an important development trend of passive radar by using multiple FM radio signals to improve detection performance, this paper presents a study on the phase compensation methods for high range resolution signal synthesis of multi-frequency passive radar. Firstly multi-FM radio received signal model is introduced and the impact of single-FM radio received signals phase on high range resolution signal synthesis is analyzed; then two phase compensation methods that are used to adaptively adjust the received signals phase are proposed to achieve the best signal synthesis effect. This paper not only elaborates in theory the internal mechanism of the methods, but also verifies multi-FM signal synthesis can distinguish adjacent targets based on the methods through simulation.
As it is an important development trend of passive radar by using multiple FM radio signals to improve detection performance, this paper presents a study on the phase compensation methods for high range resolution signal synthesis of multi-frequency passive radar. Firstly multi-FM radio received signal model is introduced and the impact of single-FM radio received signals phase on high range resolution signal synthesis is analyzed; then two phase compensation methods that are used to adaptively adjust the received signals phase are proposed to achieve the best signal synthesis effect. This paper not only elaborates in theory the internal mechanism of the methods, but also verifies multi-FM signal synthesis can distinguish adjacent targets based on the methods through simulation.
2014, 36(8): 1992-1998.
doi: 10.3724/SP.J.1146.2014.00107
Abstract:
For the space-borne timekeeping system, the relativistic effect has an important impact on the clock measurement accuracy. This paper introduces the relativistic theory into the research on wide range time-space transfer due to clock measurement error. The components and magnitudes of various relativistic effects for clocks are derived. Clock frequency adjusting and real-time periodic compensating are confirmed useful to eliminate relativistic effects. For time synchronization and transfer on common Earth-orbiting satellites, aircrafts and moon probes, an interval frequency adjustment, an improved error-restrict interval frequency adjustment and a leap seconds method are proposed to restrict secular relativistic effect errors when the clocks lack of frequency adjustable resolution, it is also proposed to predict periodic errors by building a cross correlation function when unable to get the clocks real-time position and velocity. The simulations show that the analysis and proposed error eliminating methods are effective, and they be used to improve clock synchronization for the future deep space explore.
For the space-borne timekeeping system, the relativistic effect has an important impact on the clock measurement accuracy. This paper introduces the relativistic theory into the research on wide range time-space transfer due to clock measurement error. The components and magnitudes of various relativistic effects for clocks are derived. Clock frequency adjusting and real-time periodic compensating are confirmed useful to eliminate relativistic effects. For time synchronization and transfer on common Earth-orbiting satellites, aircrafts and moon probes, an interval frequency adjustment, an improved error-restrict interval frequency adjustment and a leap seconds method are proposed to restrict secular relativistic effect errors when the clocks lack of frequency adjustable resolution, it is also proposed to predict periodic errors by building a cross correlation function when unable to get the clocks real-time position and velocity. The simulations show that the analysis and proposed error eliminating methods are effective, and they be used to improve clock synchronization for the future deep space explore.
2014, 36(8): 1999-2004.
doi: 10.3724/SP.J.1146.2013.01473
Abstract:
In the terahertz band, most targets are in the scope of near field. The computation method of the far field RCS is not applicative, therefore, the computational formula of the near-field Radar Cross Section (RCS) is deduced. The occlusion judgment consumes a long time in the Physical Optics (PO) method because of the huge amount of surface elements, and the computational error is unacceptable in the GRaphical-Electromagnetic COmputing (GRECO) method using pixels as the unit of calculation. To deal with the above problem in the PO method and the GRECO method, based on the computational parameters pretreatment, a fast computational method of near field RCS using surface element as the calculating unit and the pixel is used as the occlusion judgment unit is proposed for the complex objects in terahertz band. The method ensures the calculation accuracy, and greatly reduces the computational complexity, and the time consumption of occlusion judging. The calculation of the near-field RCS produced by a square plate, a sphere and a satellite is studied in terahertz band. The results show that this method is efficient and accurate.
In the terahertz band, most targets are in the scope of near field. The computation method of the far field RCS is not applicative, therefore, the computational formula of the near-field Radar Cross Section (RCS) is deduced. The occlusion judgment consumes a long time in the Physical Optics (PO) method because of the huge amount of surface elements, and the computational error is unacceptable in the GRaphical-Electromagnetic COmputing (GRECO) method using pixels as the unit of calculation. To deal with the above problem in the PO method and the GRECO method, based on the computational parameters pretreatment, a fast computational method of near field RCS using surface element as the calculating unit and the pixel is used as the occlusion judgment unit is proposed for the complex objects in terahertz band. The method ensures the calculation accuracy, and greatly reduces the computational complexity, and the time consumption of occlusion judging. The calculation of the near-field RCS produced by a square plate, a sphere and a satellite is studied in terahertz band. The results show that this method is efficient and accurate.
2014, 36(8): 2005-2009.
doi: 10.3724/SP.J.1146.2013.01428
Abstract:
In order to rapidly and accurately solve the radio wave propagation problems in a large scale complex electromagnetic environment with key targets, a subgridding model of the Parabolic Equation (PE) method based on the non-uniform mesh technology is presented with the detailed description on how to construct specificaly this model. The high efficiency of subgridding technique is verified by computing a complex electromagnetic environment case with a strong scattering object. The results show that the subgridding technique for the parabolic equation can improve the computational speed by 4.57 times and decrease the grid number by 86.64% as compared with the fine grid. It has a higher precision in comparition with the non-uniform mesh, demonstrating that the subgridding model can significantly enhance the simulation efficiency in solving the parabolic equation.
In order to rapidly and accurately solve the radio wave propagation problems in a large scale complex electromagnetic environment with key targets, a subgridding model of the Parabolic Equation (PE) method based on the non-uniform mesh technology is presented with the detailed description on how to construct specificaly this model. The high efficiency of subgridding technique is verified by computing a complex electromagnetic environment case with a strong scattering object. The results show that the subgridding technique for the parabolic equation can improve the computational speed by 4.57 times and decrease the grid number by 86.64% as compared with the fine grid. It has a higher precision in comparition with the non-uniform mesh, demonstrating that the subgridding model can significantly enhance the simulation efficiency in solving the parabolic equation.
2014, 36(8): 2010-2015.
doi: 10.3724/SP.J.1146.2013.01460
Abstract:
In deep-submicron Integrated Circuit (IC) design regime, the portion of leakage power consumption increases rapidly, therefore, leakage power optimization becomes a crucial part of circuit design flow. This paper proposes a mixed method of leakage optimization for gate-level netlist. The proposed method combines integer programming and heuristic algorithm to optimize leakage power at the cost of decreased timing slack. It starts at a given timing feasible design and finds alternative cell for each gate in the netlist with optimal power-delay sensitivity, then assigns alternative cell to individual gate during a levelized traverse on netlist according to specific rules. Finally, the proposed method performs a path-based timing recovery phase to fix timing violations. The entire flow iteratively converts timing slack to power-saving until no improvements could be gained. The benchmark results shows that our the proposed method achieves 10% on average, maximum 26% leakage power reduction while timing violation is confined within 5 ps.
In deep-submicron Integrated Circuit (IC) design regime, the portion of leakage power consumption increases rapidly, therefore, leakage power optimization becomes a crucial part of circuit design flow. This paper proposes a mixed method of leakage optimization for gate-level netlist. The proposed method combines integer programming and heuristic algorithm to optimize leakage power at the cost of decreased timing slack. It starts at a given timing feasible design and finds alternative cell for each gate in the netlist with optimal power-delay sensitivity, then assigns alternative cell to individual gate during a levelized traverse on netlist according to specific rules. Finally, the proposed method performs a path-based timing recovery phase to fix timing violations. The entire flow iteratively converts timing slack to power-saving until no improvements could be gained. The benchmark results shows that our the proposed method achieves 10% on average, maximum 26% leakage power reduction while timing violation is confined within 5 ps.
2014, 36(8): 2016-2022.
doi: 10.3724/SP.J.1146.2013.02005
Abstract:
With the development of semiconductor technology, integration of an increasing number of IP (Intellectual Properties) cores into a single SoC (System-on-Chip) becomes feasible. The IP cores are connected via a bus, thus the preemption of bus among IP cores degrades the performance of SoC. Efficient bus arbiters can deal with the contentions and conflicts caused by the preemption of bus, and in this way the performance of SoC is improved. An improved high-speed lottery bus arbiter is proposed. The lottery decision mechanism deploys four-phase dual-rail protocol rather than clock to avoid the loss of tickets, and it utilizes cross parallel working manner of asynchronous pipeline to improve the working speed. In the NINP (NonIdling and NonPreemptive) model, simulations and verifications are made on Xilinx Virtex5 of 65 nm CMOS. The results show that compared with the commonly-used lottery arbiter and adaptive dynamic lottery arbiter, the proposed arbiter is better in output bandwidth allocation and can avoid starvation and monopolization of bus. Furthermore, the working speed increases by over 49.2% and it has advantages in power consumption. Thus it can be applied to multi-core SoC, which has requirements in working speed.
With the development of semiconductor technology, integration of an increasing number of IP (Intellectual Properties) cores into a single SoC (System-on-Chip) becomes feasible. The IP cores are connected via a bus, thus the preemption of bus among IP cores degrades the performance of SoC. Efficient bus arbiters can deal with the contentions and conflicts caused by the preemption of bus, and in this way the performance of SoC is improved. An improved high-speed lottery bus arbiter is proposed. The lottery decision mechanism deploys four-phase dual-rail protocol rather than clock to avoid the loss of tickets, and it utilizes cross parallel working manner of asynchronous pipeline to improve the working speed. In the NINP (NonIdling and NonPreemptive) model, simulations and verifications are made on Xilinx Virtex5 of 65 nm CMOS. The results show that compared with the commonly-used lottery arbiter and adaptive dynamic lottery arbiter, the proposed arbiter is better in output bandwidth allocation and can avoid starvation and monopolization of bus. Furthermore, the working speed increases by over 49.2% and it has advantages in power consumption. Thus it can be applied to multi-core SoC, which has requirements in working speed.
2014, 36(8): 2023-2027.
doi: 10.3724/SP.J.1146.2013.01438
Abstract:
A modeling and target detection algorithm based on adaptive adjustmentK- for Mixture Gaussian background is proposed for complex scenes with non-stationary background. The Mixture Gaussian Model (GMM) is applied to learn the distribution of per-pixel in the temporal domain, then a method is constructed for adaptively adjusting the number K of Gaussian components, and the number K will be added, deleted, or merged with similar Gaussian components according to different situation. Furthermore, two new parameters are introduced in the adaptive parameter model, and the parameter is adaptively adjusted according to the actual situation, which assures that the background modeling and target detection real-time changes with the pixel. The property of real-time and accuracy reduces the loss of information for moving target and improves the robustness and convergence. Experimental results show that the algorithm responses rapidly when the scene changes in the sequence of video with many uncertain factors, and realizes adaptive background modeling and accurate target detection.
A modeling and target detection algorithm based on adaptive adjustmentK- for Mixture Gaussian background is proposed for complex scenes with non-stationary background. The Mixture Gaussian Model (GMM) is applied to learn the distribution of per-pixel in the temporal domain, then a method is constructed for adaptively adjusting the number K of Gaussian components, and the number K will be added, deleted, or merged with similar Gaussian components according to different situation. Furthermore, two new parameters are introduced in the adaptive parameter model, and the parameter is adaptively adjusted according to the actual situation, which assures that the background modeling and target detection real-time changes with the pixel. The property of real-time and accuracy reduces the loss of information for moving target and improves the robustness and convergence. Experimental results show that the algorithm responses rapidly when the scene changes in the sequence of video with many uncertain factors, and realizes adaptive background modeling and accurate target detection.
2014, 36(8): 2028-2032.
doi: 10.3724/SP.J.1146.2013.01551
Abstract:
In order to realize highly accurate and stable clock signal extraction using Fiber Optical Parametric Oscillators (FOPOs), the idler-power-based automatic feedback control experiment is implemented for the first time. The automatic feedback control unit consists of idler acquisition module, signal processing module for automatic control, and delay line drive module. By means of the automatically feedback-controlled FOPO, a high-quality optical clock signal can be extracted from the 12.5 Gb/s Return-to-Zero (RZ) data sequence and the RMS phase jitter is 0.92 ps, much better than the results obtained in the previous manual feedback control system.
In order to realize highly accurate and stable clock signal extraction using Fiber Optical Parametric Oscillators (FOPOs), the idler-power-based automatic feedback control experiment is implemented for the first time. The automatic feedback control unit consists of idler acquisition module, signal processing module for automatic control, and delay line drive module. By means of the automatically feedback-controlled FOPO, a high-quality optical clock signal can be extracted from the 12.5 Gb/s Return-to-Zero (RZ) data sequence and the RMS phase jitter is 0.92 ps, much better than the results obtained in the previous manual feedback control system.
