Email alert
2011 Vol. 33, No. 4
Display Method:
2011, 33(4): 763-768.
doi: 10.3724/SP.J.1146.2010.00839
Abstract:
Fast and efficient segmentation of rigid or stable object in multi-view images is still a unsolved problem. In this paper, the problem is formulated and the relationships between it and traditional min-cut based segmentation problems are also deduced with Graph representation. To minimize the energy function, a novel algorithm named Interactive Transductive Co-Segmentation (ITC-Seg) is proposed. In ITC-Seg, the function is divided into two sub problems which are solved with graph cuts and spectral segmentation methods. Moreover, propagation, filter, voting methods are introduced into the iteration between multi-view images segmentation and qusi-sparse 3D points segmentation, they are used to combine the sub problems with a global label consistent constrains. Finally, the experiments in several images show the error rate of ITC-Seg is 3.4%, discussions and future improvements of the method are also given.
Fast and efficient segmentation of rigid or stable object in multi-view images is still a unsolved problem. In this paper, the problem is formulated and the relationships between it and traditional min-cut based segmentation problems are also deduced with Graph representation. To minimize the energy function, a novel algorithm named Interactive Transductive Co-Segmentation (ITC-Seg) is proposed. In ITC-Seg, the function is divided into two sub problems which are solved with graph cuts and spectral segmentation methods. Moreover, propagation, filter, voting methods are introduced into the iteration between multi-view images segmentation and qusi-sparse 3D points segmentation, they are used to combine the sub problems with a global label consistent constrains. Finally, the experiments in several images show the error rate of ITC-Seg is 3.4%, discussions and future improvements of the method are also given.
2011, 33(4): 769-774.
doi: 10.3724/SP.J.1146.2010.01003
Abstract:
This paper first discusses the relationship of Principal Component Analysis (PCA) and 2-Dimensional PCA (2DPCA). For 2DPCA eliminating the some covariance information which can be useful for recognition, and PCAs small sample size problem, an algorithm of feature extraction of face based on the Weighted Variation 2DPCA (WV2DPCA) is proposed. Three sub-parts of the face features are extracted respectively in the method of the variation of 2DPCA, and then are classified according to the weight and the Nearest neighbor theory. The experiments on both of ORL and YALE face bases show improvement in recognition accuracy, fewer coefficients and recognition time over 2DPCA, and this algorithm is also superior to the traditional eigenfaces, ICA and Kernel Eigenfaces in terms of the recognition accuracy.
This paper first discusses the relationship of Principal Component Analysis (PCA) and 2-Dimensional PCA (2DPCA). For 2DPCA eliminating the some covariance information which can be useful for recognition, and PCAs small sample size problem, an algorithm of feature extraction of face based on the Weighted Variation 2DPCA (WV2DPCA) is proposed. Three sub-parts of the face features are extracted respectively in the method of the variation of 2DPCA, and then are classified according to the weight and the Nearest neighbor theory. The experiments on both of ORL and YALE face bases show improvement in recognition accuracy, fewer coefficients and recognition time over 2DPCA, and this algorithm is also superior to the traditional eigenfaces, ICA and Kernel Eigenfaces in terms of the recognition accuracy.
2011, 33(4): 775-780.
doi: 10.3724/SP.J.1146.2010.00885
Abstract:
Hip angle is a major parameter in gait analysis while gait analysis plays important role in healthcare, animation and other applications. Accurate estimation of hip angle using wearable inertial sensors in ambulatory environment remains a challenge, this is mainly because (1) the non-linear nature of thigh movement has not been well addressed, and (2) the variation of micro-inertial sensor measurement noise has not been studied yet. We propose to use Hybrid Dynamic Bayesian Network (HDBN) and multiple motion models and multiple noise models to model the non-linear hip angle dynamics and variation of measurement noise levels. Gaussian Particle Filter (GPF) is employed to estimate the hip angle during gait cycles from the measurements of accelerometers that are attached to the thighs. The experimental results show that the proposed method can achieve significant accuracy improvement over the previous work on the ambulatory hip angle estimation.
Hip angle is a major parameter in gait analysis while gait analysis plays important role in healthcare, animation and other applications. Accurate estimation of hip angle using wearable inertial sensors in ambulatory environment remains a challenge, this is mainly because (1) the non-linear nature of thigh movement has not been well addressed, and (2) the variation of micro-inertial sensor measurement noise has not been studied yet. We propose to use Hybrid Dynamic Bayesian Network (HDBN) and multiple motion models and multiple noise models to model the non-linear hip angle dynamics and variation of measurement noise levels. Gaussian Particle Filter (GPF) is employed to estimate the hip angle during gait cycles from the measurements of accelerometers that are attached to the thighs. The experimental results show that the proposed method can achieve significant accuracy improvement over the previous work on the ambulatory hip angle estimation.
2011, 33(4): 781-786.
doi: 10.3724/SP.J.1146.2010.00763
Abstract:
Based on the estimated polarimetric coherence matrices of the three different scattering mechanisms by Freeman decomposition, the polarimetric interferometric cross-coherence matrices could be simply modeled as a function of the vegetation height, the extinction coefficient and the topographic phase. Based on this model, a new optimization model is established. The target function of this new optimization model is the difference between the computed results of polarimetric interferometric cross-coherence matrices and their estimated results, the variables are the vegetation parameters. A new vegetation parameters inversion method based on the Freeman decomposition is proposed. Compared to three-stage inversion process, this new method does not need to estimate the volume correlation coefficient, so it could avoid the error of the estimation of the volume correlation coefficient and more importantly this new method supplies a new approach to the inversion of the vegetation height. The simulated PolInSAR data by the PolSARpro software is processed to validate the novel method.
Based on the estimated polarimetric coherence matrices of the three different scattering mechanisms by Freeman decomposition, the polarimetric interferometric cross-coherence matrices could be simply modeled as a function of the vegetation height, the extinction coefficient and the topographic phase. Based on this model, a new optimization model is established. The target function of this new optimization model is the difference between the computed results of polarimetric interferometric cross-coherence matrices and their estimated results, the variables are the vegetation parameters. A new vegetation parameters inversion method based on the Freeman decomposition is proposed. Compared to three-stage inversion process, this new method does not need to estimate the volume correlation coefficient, so it could avoid the error of the estimation of the volume correlation coefficient and more importantly this new method supplies a new approach to the inversion of the vegetation height. The simulated PolInSAR data by the PolSARpro software is processed to validate the novel method.
2011, 33(4): 787-791.
doi: 10.3724/SP.J.1146.2010.00681
Abstract:
Based on the traditional DBS (Doppler Beam Sharpening) method and the least squares linear fitting algorithm, a new high resolution DBS imaging algorithm is delivered in this paper. The coherent processing number is discussed at first, and the range migration is corrected later. Eventually, as the focus processing of SAR, the azimuth dechirping technique is presented to improve the azimuth resolution. The processing results of real airborne radar data prove the effectiveness of the algorithm.
Based on the traditional DBS (Doppler Beam Sharpening) method and the least squares linear fitting algorithm, a new high resolution DBS imaging algorithm is delivered in this paper. The coherent processing number is discussed at first, and the range migration is corrected later. Eventually, as the focus processing of SAR, the azimuth dechirping technique is presented to improve the azimuth resolution. The processing results of real airborne radar data prove the effectiveness of the algorithm.
2011, 33(4): 792-797.
doi: 10.3724/SP.J.1146.2010.00812
Abstract:
The MOtion COmpensation (MOCO) of Light Of Sight (LOS) error is very important in low-frequency Ultra Wide-Band Synthetic Aperture Radar (UWB SAR). In this paper, the second-order MOCO is studied which compensate the range-variant component of LOS error. In second-order MOCO, the assumption that there is no effect of range-variant LOS error on the Range Cell Migration Correction (RCMC) is supposed, which is not satisfied in UWB SAR. This paper analyzes the application of the second-order MOCO in low-frequency UWB SAR. Three methods which can be applied to compensate the range-variant LOS error are presented. First method implements the second-order MOCO after RCMC, second method implements it before RCMC, and the last method implements it in sub-band which is split along range direction. The three methods are compared and analyzed in this paper. Finally, the effectiveness of theoretical analysis is validated with simulation and real low-frequency UWB SAR data.
The MOtion COmpensation (MOCO) of Light Of Sight (LOS) error is very important in low-frequency Ultra Wide-Band Synthetic Aperture Radar (UWB SAR). In this paper, the second-order MOCO is studied which compensate the range-variant component of LOS error. In second-order MOCO, the assumption that there is no effect of range-variant LOS error on the Range Cell Migration Correction (RCMC) is supposed, which is not satisfied in UWB SAR. This paper analyzes the application of the second-order MOCO in low-frequency UWB SAR. Three methods which can be applied to compensate the range-variant LOS error are presented. First method implements the second-order MOCO after RCMC, second method implements it before RCMC, and the last method implements it in sub-band which is split along range direction. The three methods are compared and analyzed in this paper. Finally, the effectiveness of theoretical analysis is validated with simulation and real low-frequency UWB SAR data.
2011, 33(4): 798-804.
doi: 10.3724/SP.J.1146.2010.00668
Abstract:
Burst mode and azimuth beam steering lead to echoes of TOPSAR mode obtain echo characteristics of both ScanSAR and spotlight mode, such as azimuth data folding in the output time domain and Doppler spectrum aliasing. Up-sampling both in the slow time and Doppler frequency domain can resolve the aliasing problem but with the unwanted result of seriously increased computational complexity and memory consumption. According to the TOPSAR raw data support in the slow Time/Frequency Domain (TFD), a new TOPSAR imaging approach combined with azimuth pre-processing and post-processing operations is proposed. The mosaiking with deramping operation approach is adopted to resolve the data aliasing problem. Furthermore, azimuth data interception is used to eliminate the disturbance signals and reduce azimuth total samples to improve the computational efficiency. Simulation results of point targets and distributed target validate the effectiveness of the presented imaging approach.
Burst mode and azimuth beam steering lead to echoes of TOPSAR mode obtain echo characteristics of both ScanSAR and spotlight mode, such as azimuth data folding in the output time domain and Doppler spectrum aliasing. Up-sampling both in the slow time and Doppler frequency domain can resolve the aliasing problem but with the unwanted result of seriously increased computational complexity and memory consumption. According to the TOPSAR raw data support in the slow Time/Frequency Domain (TFD), a new TOPSAR imaging approach combined with azimuth pre-processing and post-processing operations is proposed. The mosaiking with deramping operation approach is adopted to resolve the data aliasing problem. Furthermore, azimuth data interception is used to eliminate the disturbance signals and reduce azimuth total samples to improve the computational efficiency. Simulation results of point targets and distributed target validate the effectiveness of the presented imaging approach.
2011, 33(4): 805-809.
doi: 10.3724/SP.J.1146.2010.00704
Abstract:
A computationally efficient Space-Time Adaptive Processing (STAP) method for clutter suppression in airborne MIMO radar is proposed. Firstly, the Doppler filtering is performed to reduce the data dimension in temporal domain. Secondly, the weight vector of the two-dimensional transmit-receive beamformer is expressed as the Kronecker product of two short weight vectors for which a bi-quadratic cost function is constructed to achieve dimension reduction in spatial domain. Finally, a bi-iterative algorithm for minimizing the bi-quadratic cost function is utilized to optimize iteratively the two weights. Simulation results indicate that the proposed method has fast convergence rate and particularly for short data records, it provides substantially better clutter suppression performance and can be carried out at a smaller computational cost than the mDT method.
A computationally efficient Space-Time Adaptive Processing (STAP) method for clutter suppression in airborne MIMO radar is proposed. Firstly, the Doppler filtering is performed to reduce the data dimension in temporal domain. Secondly, the weight vector of the two-dimensional transmit-receive beamformer is expressed as the Kronecker product of two short weight vectors for which a bi-quadratic cost function is constructed to achieve dimension reduction in spatial domain. Finally, a bi-iterative algorithm for minimizing the bi-quadratic cost function is utilized to optimize iteratively the two weights. Simulation results indicate that the proposed method has fast convergence rate and particularly for short data records, it provides substantially better clutter suppression performance and can be carried out at a smaller computational cost than the mDT method.
2011, 33(4): 810-815.
doi: 10.3724/SP.J.1146.2010.00648
Abstract:
In this paper, the Doppler centroid frequency and the Doppler frequency rate formulas of spaceborne SAR with arbitrary orbit altitude are derived, which provides an effective method for Doppler properties study of MEO SAR and HEO SAR. Based on these formulas, the relation of Doppler centroid frequency between GEO SAR and LEO SAR is analyzed. The change of spaceborne SAR Doppler frequency rate with the orbit altitude, as well as the influence of earth rotation on the Doppler frequency rate of LEO SAR and that of GEO SAR are studied. The properties of Doppler centroid frequency and Doppler frequency rate of GEO SAR is discussed in detail. The validities of formulas and the analysis of Doppler properties are demonstrated by the numeric simulations.
In this paper, the Doppler centroid frequency and the Doppler frequency rate formulas of spaceborne SAR with arbitrary orbit altitude are derived, which provides an effective method for Doppler properties study of MEO SAR and HEO SAR. Based on these formulas, the relation of Doppler centroid frequency between GEO SAR and LEO SAR is analyzed. The change of spaceborne SAR Doppler frequency rate with the orbit altitude, as well as the influence of earth rotation on the Doppler frequency rate of LEO SAR and that of GEO SAR are studied. The properties of Doppler centroid frequency and Doppler frequency rate of GEO SAR is discussed in detail. The validities of formulas and the analysis of Doppler properties are demonstrated by the numeric simulations.
2011, 33(4): 816-822.
doi: 10.3724/SP.J.1146.2010.00678
Abstract:
In this paper, based on polarization characteristics decomposition of full-polarization information processing, a novel method of detecting low observable targets within sea clutter at low grazing angle is proposed for the first time. Through analyzing the real full-polarization data, the polarization characteristic of sea clutter, described before, has been verified. The differences of polarization characteristics, after polarization characteristics decomposition, between pure sea clutter and the one including the small target have been compared, according to which a novel polarization characteristic, named Distance Between Entropy and Anisotropy (DBEA), is proposed. The detection result can be obtained after the new method based on the scattering mechanism difference of DBEA. It has been proved that the new way has good detection performance and strong robustness with more times random test and under more sea state levels.
In this paper, based on polarization characteristics decomposition of full-polarization information processing, a novel method of detecting low observable targets within sea clutter at low grazing angle is proposed for the first time. Through analyzing the real full-polarization data, the polarization characteristic of sea clutter, described before, has been verified. The differences of polarization characteristics, after polarization characteristics decomposition, between pure sea clutter and the one including the small target have been compared, according to which a novel polarization characteristic, named Distance Between Entropy and Anisotropy (DBEA), is proposed. The detection result can be obtained after the new method based on the scattering mechanism difference of DBEA. It has been proved that the new way has good detection performance and strong robustness with more times random test and under more sea state levels.
2011, 33(4): 823-830.
doi: 10.3724/SP.J.1146.2010.00486
Abstract:
A new moving target detection algorithm is proposed based on the joint fractal properties discriminant of sea clutter in FRactional Fourier Transform (FRFT) domain. At first, statistical characteristic of sea clutter data in FRFT domain is analyzed and simulations of fractal curves in different polarizations are conducted, which indicates the self-similarity feature. Then, determination method of fractal parameters and scale-invariant interval is given and influence of transform order on the estimation of fractal parameters is also discussed. Finally, differences of fractal dimension and intercept in FRFT domain, which are calculated from adjacent range bin or time series of radar echo, can be used as test statistic. Real sea clutter in different polarizations is used for verification and the results present that the proposed algorithm has good performance for weak moving target detection and can also give high estimation accuracy of moving conditions.
A new moving target detection algorithm is proposed based on the joint fractal properties discriminant of sea clutter in FRactional Fourier Transform (FRFT) domain. At first, statistical characteristic of sea clutter data in FRFT domain is analyzed and simulations of fractal curves in different polarizations are conducted, which indicates the self-similarity feature. Then, determination method of fractal parameters and scale-invariant interval is given and influence of transform order on the estimation of fractal parameters is also discussed. Finally, differences of fractal dimension and intercept in FRFT domain, which are calculated from adjacent range bin or time series of radar echo, can be used as test statistic. Real sea clutter in different polarizations is used for verification and the results present that the proposed algorithm has good performance for weak moving target detection and can also give high estimation accuracy of moving conditions.
2011, 33(4): 831-837.
doi: 10.3724/SP.J.1146.2010.00750
Abstract:
Airborne forward-looking SAR based on array technology can supply high-resolution images of the area with respect to forward direction of the flight path. However, micro-motion of the array antenna may affect imaging. Extraction method of micro-motion characteristic parameters based on time-frequency analysis is proposed, and influence of array antenna micro-motion on forward-looking SAR imaging and the compensation method are investigated. Firstly, micro-motion model of array antenna is established based on antenna characteristic of forward-looking SAR, and compensation method of array antenna deformation is analyzed. Then, time-frequency characteristic of array antenna micro-motion is studied based on time-frequency analysis method. Finally, imaging simulation of point target in the condition of antenna micro-motion is done. Imaging results before and after compensation are compared combined with imaging characteristic. Results indicate that the established model and proposed method are correct and valid.
Airborne forward-looking SAR based on array technology can supply high-resolution images of the area with respect to forward direction of the flight path. However, micro-motion of the array antenna may affect imaging. Extraction method of micro-motion characteristic parameters based on time-frequency analysis is proposed, and influence of array antenna micro-motion on forward-looking SAR imaging and the compensation method are investigated. Firstly, micro-motion model of array antenna is established based on antenna characteristic of forward-looking SAR, and compensation method of array antenna deformation is analyzed. Then, time-frequency characteristic of array antenna micro-motion is studied based on time-frequency analysis method. Finally, imaging simulation of point target in the condition of antenna micro-motion is done. Imaging results before and after compensation are compared combined with imaging characteristic. Results indicate that the established model and proposed method are correct and valid.
2011, 33(4): 838-843.
doi: 10.3724/SP.J.1146.2010.00730
Abstract:
The paper proposes a scheme for the realization of multi-beam Synthetic Aperture Sonar (SAS) based on the multi-beam bathymetry and aperture synthetic principle, designs a new multiple receiver array structure, deduces correlated math model, and simulates the dot-by-dot imaging algorithm according to the array structure. Computer simulation results show that: compared with multi-beam bathymetry sonar, the new scheme can remarkably improve the along-track resolution, and compared with conventional SAS, the new scheme can get more information of the target and enhance the imaging quality.
The paper proposes a scheme for the realization of multi-beam Synthetic Aperture Sonar (SAS) based on the multi-beam bathymetry and aperture synthetic principle, designs a new multiple receiver array structure, deduces correlated math model, and simulates the dot-by-dot imaging algorithm according to the array structure. Computer simulation results show that: compared with multi-beam bathymetry sonar, the new scheme can remarkably improve the along-track resolution, and compared with conventional SAS, the new scheme can get more information of the target and enhance the imaging quality.
2011, 33(4): 844-848.
doi: 10.3724/SP.J.1146.2010.00807
Abstract:
The hyperchaotic system, composed of high and slow frequency part, is robust to strong noise. If the conventional synchronization algorithms are used in this system, the convergence rate of synchronization is slow and synchronization performances are sensitive to the parameters of response system. To resolve the problems, the synchronization algorithm using nonlinear feedback control is proposed, according to Hurwitz stability theory, designing a nonlinear controller can make the real parts of the eigenvalues of the error equations Jacobian matrix negative and the absolutes of the eigenvalues large. So the convergence rate of synchronization is faster than that of conventional synchronization, besides, it has the advantage of parameters insensitivity. Finally, numerical simulations are given to verify the effectiveness of the proposed methods.
The hyperchaotic system, composed of high and slow frequency part, is robust to strong noise. If the conventional synchronization algorithms are used in this system, the convergence rate of synchronization is slow and synchronization performances are sensitive to the parameters of response system. To resolve the problems, the synchronization algorithm using nonlinear feedback control is proposed, according to Hurwitz stability theory, designing a nonlinear controller can make the real parts of the eigenvalues of the error equations Jacobian matrix negative and the absolutes of the eigenvalues large. So the convergence rate of synchronization is faster than that of conventional synchronization, besides, it has the advantage of parameters insensitivity. Finally, numerical simulations are given to verify the effectiveness of the proposed methods.
2011, 33(4): 849-853.
doi: 10.3724/SP.J.1146.2010.00731
Abstract:
Honest-sender commitments were introduced to construct non-interactive and non-malleable commitments. In this paper, two honest-sender commitments are given: The first one is honest-sender statistically binding and the second one is perfectly hiding. Neither of them is a commitment scheme, which implies that honest-sender commitment is a weaker cryptographic primitive than commitment scheme. In addition, this paper defines honest-receiver commitment and gives two constructions: The first one has the statistically binding property and the second has the honest-receiver perfectly hiding property. Both of them are not commitment scheme, which implies that honest-receiver commitment is a weaker cryptographic primitive than commitment scheme.
Honest-sender commitments were introduced to construct non-interactive and non-malleable commitments. In this paper, two honest-sender commitments are given: The first one is honest-sender statistically binding and the second one is perfectly hiding. Neither of them is a commitment scheme, which implies that honest-sender commitment is a weaker cryptographic primitive than commitment scheme. In addition, this paper defines honest-receiver commitment and gives two constructions: The first one has the statistically binding property and the second has the honest-receiver perfectly hiding property. Both of them are not commitment scheme, which implies that honest-receiver commitment is a weaker cryptographic primitive than commitment scheme.
2011, 33(4): 854-857.
doi: 10.3724/SP.J.1146.2010.00837
Abstract:
SP (Substitution Permutation) structure is often used in block ciphers. This paper provides a method which could construct high probability differential trails and linear trails by using characteristic vectors of the diffusion layer. By this method some differential trails of ARIA can be constructed, these trails could reach probability2-168 for 6 rounds. And for 6 rounds reduced ARIA, who only employs a single S box, some differential trails can be got which could reach the highest probability 2-144. The results show that the SP cipher designers should choose those diffusion layers with fewer characteristic vectors as possible. And diffusion layers should never have low weight characteristic vectors. Additionally, the characteristic value as well as the count value of quasi-involution MDS matrices and cyclic shift matrices are provided.
SP (Substitution Permutation) structure is often used in block ciphers. This paper provides a method which could construct high probability differential trails and linear trails by using characteristic vectors of the diffusion layer. By this method some differential trails of ARIA can be constructed, these trails could reach probability2-168 for 6 rounds. And for 6 rounds reduced ARIA, who only employs a single S box, some differential trails can be got which could reach the highest probability 2-144. The results show that the SP cipher designers should choose those diffusion layers with fewer characteristic vectors as possible. And diffusion layers should never have low weight characteristic vectors. Additionally, the characteristic value as well as the count value of quasi-involution MDS matrices and cyclic shift matrices are provided.
2011, 33(4): 858-863.
doi: 10.3724/SP.J.1146.2010.00708
Abstract:
A novel retransmission approach based on Opportunistic Network Coding (ONC) is proposed to improve the transmission efficiency in wireless broadcast scenario. The ONC is applied to the retransmission of lost packets, and an efficient combination strategy is introduced to generate the retransmission packet. According to the packet loss information of terminals, a Hash table is built, on the basis of which the lost packets are selected in a certain way to generate the retransmission packet rapidly. So the proposed approach can improve the performance of retransmission and reduce the complexity effectively. The simulation results show that the proposed approach can reduce the number of retransmission, and improve the efficiency of packet retransmission and retrievement compared with existing approaches.
A novel retransmission approach based on Opportunistic Network Coding (ONC) is proposed to improve the transmission efficiency in wireless broadcast scenario. The ONC is applied to the retransmission of lost packets, and an efficient combination strategy is introduced to generate the retransmission packet. According to the packet loss information of terminals, a Hash table is built, on the basis of which the lost packets are selected in a certain way to generate the retransmission packet rapidly. So the proposed approach can improve the performance of retransmission and reduce the complexity effectively. The simulation results show that the proposed approach can reduce the number of retransmission, and improve the efficiency of packet retransmission and retrievement compared with existing approaches.
2011, 33(4): 864-868.
doi: 10.3724/SP.J.1146.2010.00134
Abstract:
Gardner bit synchronization method in single carrier is applied to OFDM system, to complete sampling rate synchronization. Gardner loop is composed of timing error estimator, loop filter and interpolation. The loop filter is replaced by linear least-square fitting which is achieved by recursion in order to reduce computational complexity. The estimation error of sampling rate can be reduced in cubic law. But there are two problems in practice when using least-square fitting. First, the sampling rate is not constant, which will change slowly by the temperature. Second, the intermediate variables will overflow in recursion. So the intermediate variables are modified periodically to limiting the time span and change the coordinate to solve the two problems.
Gardner bit synchronization method in single carrier is applied to OFDM system, to complete sampling rate synchronization. Gardner loop is composed of timing error estimator, loop filter and interpolation. The loop filter is replaced by linear least-square fitting which is achieved by recursion in order to reduce computational complexity. The estimation error of sampling rate can be reduced in cubic law. But there are two problems in practice when using least-square fitting. First, the sampling rate is not constant, which will change slowly by the temperature. Second, the intermediate variables will overflow in recursion. So the intermediate variables are modified periodically to limiting the time span and change the coordinate to solve the two problems.
2011, 33(4): 869-873.
doi: 10.3724/SP.J.1146.2010.00830
Abstract:
It is already theoretically presented in previous work that multiple cognitive relay improves the throughput of cognitive networks by increasing diversity in both spatial and temporary domain. According to observation and analysis, huge amounts of redundancies exist when implementing Decode-and-Forward (DF) in a distributed relay network. To solve the issue, a so-called code-assist scheme is proposed in this paper to eliminate redundancies. By such a scheme, network coding is integrated with multiple cognitive relay. It is proved that the achievable throughput of cognitive users approaches to its theoretical upper-bound under the scheme. Moreover, this scheme also helps to decrease the effective buffer size for relay. Simulation results show that the scheme is promising and efficient.
It is already theoretically presented in previous work that multiple cognitive relay improves the throughput of cognitive networks by increasing diversity in both spatial and temporary domain. According to observation and analysis, huge amounts of redundancies exist when implementing Decode-and-Forward (DF) in a distributed relay network. To solve the issue, a so-called code-assist scheme is proposed in this paper to eliminate redundancies. By such a scheme, network coding is integrated with multiple cognitive relay. It is proved that the achievable throughput of cognitive users approaches to its theoretical upper-bound under the scheme. Moreover, this scheme also helps to decrease the effective buffer size for relay. Simulation results show that the scheme is promising and efficient.
2011, 33(4): 874-879.
doi: 10.3724/SP.J.1146.2010.00692
Abstract:
In this paper , the evolution equations of solitons performance parameters versus distance are deduced by the use of the variational approach under the condition of both the higher-order effects and interaction between solitons for the first time. Based on this equation, a stable operation condition is revealed from theoretical analysis by using a linearization perturbation method. It is discovered that the stable transmission of the Gaussian quasi-soliton system with zero frequency shift at bit rates up to 160 Gb/s can be realized. A 160 Gb/s high-speed and long-haul soliton transmission system is designed and its performance is numerically simulated. The simulation results are in well accordance with the theoretical analysis.
In this paper , the evolution equations of solitons performance parameters versus distance are deduced by the use of the variational approach under the condition of both the higher-order effects and interaction between solitons for the first time. Based on this equation, a stable operation condition is revealed from theoretical analysis by using a linearization perturbation method. It is discovered that the stable transmission of the Gaussian quasi-soliton system with zero frequency shift at bit rates up to 160 Gb/s can be realized. A 160 Gb/s high-speed and long-haul soliton transmission system is designed and its performance is numerically simulated. The simulation results are in well accordance with the theoretical analysis.
2011, 33(4): 880-884.
doi: 10.3724/SP.J.1146.2010.00930
Abstract:
The Channel State Information (CSI) is not taken into consideration in Multi-Hop Alamouti Amplify and Forward (AAF-MH) cooperation protocol. For this issue, Multi-Hop Alamouti Amplify and Forward with Phase Rotation (AAF-MH-PR) cooperative scheme is proposed, in which the effective channel matrix is changed by applying a proper rotation at transmitted signals of partial relay. Theoretical analysis shows that the reliability of AAF-MH-PR is significantly better than that of AAF-MH. Simulation demonstrats the validity of theoretical analysis and the proposed scheme provides a 5 dB gain at a Symbol Error Rate (SER) of10-5 over AAF-MH with only 1 bit feedback amount.
The Channel State Information (CSI) is not taken into consideration in Multi-Hop Alamouti Amplify and Forward (AAF-MH) cooperation protocol. For this issue, Multi-Hop Alamouti Amplify and Forward with Phase Rotation (AAF-MH-PR) cooperative scheme is proposed, in which the effective channel matrix is changed by applying a proper rotation at transmitted signals of partial relay. Theoretical analysis shows that the reliability of AAF-MH-PR is significantly better than that of AAF-MH. Simulation demonstrats the validity of theoretical analysis and the proposed scheme provides a 5 dB gain at a Symbol Error Rate (SER) of10-5 over AAF-MH with only 1 bit feedback amount.
2011, 33(4): 885-890.
doi: 10.3724/SP.J.1146.2010.00836
Abstract:
It has been proved that many popular link quality prediction algorithms perform much more poorly when they are applied to real environments. To address this problem, a low-complexity, on-line link quality prediction mechanism is constructed which can be applied to reliable routing protocol design. The algorithm does not require the use of any deterministic propagation model or mobility model; it relies simply on Received Signal Strength (RSS) to estimate link quality. The nodes monitor and store the links signal strength values to their communicating neighbors in order to obtain a time series of RSS measurements. Local linear kernel smoothing method and moving window local polynomial prediction method in time series analysis are introduced to construct an efficient link quality prediction mechanism, which provides significant cross-layer information for the design of mobility-adaptive proactive route repair scheme. Simulation results indicate that the proposed algorithm can provide high accuracy of link quality prediction, resulting in improved route stability and network performance.
It has been proved that many popular link quality prediction algorithms perform much more poorly when they are applied to real environments. To address this problem, a low-complexity, on-line link quality prediction mechanism is constructed which can be applied to reliable routing protocol design. The algorithm does not require the use of any deterministic propagation model or mobility model; it relies simply on Received Signal Strength (RSS) to estimate link quality. The nodes monitor and store the links signal strength values to their communicating neighbors in order to obtain a time series of RSS measurements. Local linear kernel smoothing method and moving window local polynomial prediction method in time series analysis are introduced to construct an efficient link quality prediction mechanism, which provides significant cross-layer information for the design of mobility-adaptive proactive route repair scheme. Simulation results indicate that the proposed algorithm can provide high accuracy of link quality prediction, resulting in improved route stability and network performance.
2011, 33(4): 891-895.
doi: 10.3724/SP.J.1146.2010.00780
Abstract:
The Received Signal Strength Indicator (RSSI) based location is easy to be affected by multipath interference in wireless sensor networks. The traditional RSSI-based location methods usually increase the number of samples to resist multipath interference while few works are done on physical layer. This paper studies to separate subchip multipath components for RSSI-based location application, the direct path can be separated among multipath by despreaded using the first sample of each chip, after calculating the direct paths power the anti-multipath interferences RSSI value can be got. The theoretical analysis of the performance of separating subchip multipath, Signal-to-Noise Ratio (SNR) losses are presented. Finally, the simulation shows the variance of RSSI value calculated by the traditional method and the proposal in this paper, the variance of the proposed method is smaller than the traditional method under multipath channel environment.
The Received Signal Strength Indicator (RSSI) based location is easy to be affected by multipath interference in wireless sensor networks. The traditional RSSI-based location methods usually increase the number of samples to resist multipath interference while few works are done on physical layer. This paper studies to separate subchip multipath components for RSSI-based location application, the direct path can be separated among multipath by despreaded using the first sample of each chip, after calculating the direct paths power the anti-multipath interferences RSSI value can be got. The theoretical analysis of the performance of separating subchip multipath, Signal-to-Noise Ratio (SNR) losses are presented. Finally, the simulation shows the variance of RSSI value calculated by the traditional method and the proposal in this paper, the variance of the proposed method is smaller than the traditional method under multipath channel environment.
2011, 33(4): 896-901.
doi: 10.3724/SP.J.1146.2010.00813
Abstract:
The time-varying Received Signal Strength (RSS) drastically degrades the indoor positioning accuracy in Wireless Local Area Network (WLAN). A new positioning algorithm based on Kernel Direct Discriminant Analysis (KDDA) and Support Vector Regression (SVR) is proposed to resolve the problem in this paper. The proposed algorithm employs KDDA to reconstruct the localization information contained in the RSS signal. The most discriminative localization features are then extracted while the redundant localization features and noise are discarded by KDDA. The extracted localization features are taken as inputs to SVR learning machine and the mapping between localization features and physical locations is established. The experimental results show that the proposed algorithm obtains significant accuracy improvement while requiring a much smaller set of RSS training data than previous methods.
The time-varying Received Signal Strength (RSS) drastically degrades the indoor positioning accuracy in Wireless Local Area Network (WLAN). A new positioning algorithm based on Kernel Direct Discriminant Analysis (KDDA) and Support Vector Regression (SVR) is proposed to resolve the problem in this paper. The proposed algorithm employs KDDA to reconstruct the localization information contained in the RSS signal. The most discriminative localization features are then extracted while the redundant localization features and noise are discarded by KDDA. The extracted localization features are taken as inputs to SVR learning machine and the mapping between localization features and physical locations is established. The experimental results show that the proposed algorithm obtains significant accuracy improvement while requiring a much smaller set of RSS training data than previous methods.
2011, 33(4): 902-907.
doi: 10.3724/SP.J.1146.2010.00751
Abstract:
To address the state explosion problem in the procedure of model checking, this paper proposes a SPIN- based modularized model checking method. The proposed method divides the original abstraction model into some modules, and verifies all the divided modules, instead of verifying the original model. In the dividing process, the semantic of original model can be completely included in the modules, and no semantic is added in the process. Consequently, the original model passes the verification of model checking if and only if all the modules pass the verification. The theoretical and experimental results show that the proposed method is valid.
To address the state explosion problem in the procedure of model checking, this paper proposes a SPIN- based modularized model checking method. The proposed method divides the original abstraction model into some modules, and verifies all the divided modules, instead of verifying the original model. In the dividing process, the semantic of original model can be completely included in the modules, and no semantic is added in the process. Consequently, the original model passes the verification of model checking if and only if all the modules pass the verification. The theoretical and experimental results show that the proposed method is valid.
2011, 33(4): 908-914.
doi: 10.3724/SP.J.1146.2010.00735
Abstract:
Based on the previous research, a new algorithm with the K-shortest path is proposed. This algorithm searches the K-shortest path for increasing K, and stops the search if one with enough bandwidth capacity is found. The search process enhances the percentage of link mapping effectively. In addition, according to the rules of node mapping in general algorithm, a new virtual network mapping algorithm with repeatable mapping over substrate nodes is put forward, in which, multiple virtual nodes in the same virtual network can be mapped to the same node in the substrate network if there is enough resource capacity. That not only cuts down some virtual link mapping processes and reduces the mapping time, but also saves the substrate link cost and makes more virtual networks mapped. Experiments show that the new algorithm performs better in mapping percentage, time and mapping revenue.
Based on the previous research, a new algorithm with the K-shortest path is proposed. This algorithm searches the K-shortest path for increasing K, and stops the search if one with enough bandwidth capacity is found. The search process enhances the percentage of link mapping effectively. In addition, according to the rules of node mapping in general algorithm, a new virtual network mapping algorithm with repeatable mapping over substrate nodes is put forward, in which, multiple virtual nodes in the same virtual network can be mapped to the same node in the substrate network if there is enough resource capacity. That not only cuts down some virtual link mapping processes and reduces the mapping time, but also saves the substrate link cost and makes more virtual networks mapped. Experiments show that the new algorithm performs better in mapping percentage, time and mapping revenue.
2011, 33(4): 915-921.
doi: 10.3724/SP.J.1146.2010.00729
Abstract:
Regional MANET which contains near space vehicles as a part is a new kind of Ad hoc network. This paper proposed a deployment scheme for such network based on Particle Swarm Optimization (PSO) algorithm. For nodes which are near space vehicles, their deployment aims at maximizing joint communication covered space of the network. One particle represents a set of positions of them and optimized deployment positions for near space vehicles can be obtained with iterative method. For controllable backbone nodes on earth surface, their deployment is executed within each groups which they belong and thus one particle represents position of one backbone node. During network initial stage, the deployment aims at maximizing the number of common nodes covered by backbone nodes and minimizing average hops from common nodes to backbone nodes. During network running stage, positions of backbone nodes are dynamically adjusted to best serve their groups. The dynamic deployment mechanism ensures backbone nodes are closest to the most heavily loaded nodes, or to the nodes that generate packets of the highest priority. Simulation results show that the proposed deployment scheme can provides better operation basis for upper layer protocols in terms of network performance metrics such as throughput, latency and control overhead.
Regional MANET which contains near space vehicles as a part is a new kind of Ad hoc network. This paper proposed a deployment scheme for such network based on Particle Swarm Optimization (PSO) algorithm. For nodes which are near space vehicles, their deployment aims at maximizing joint communication covered space of the network. One particle represents a set of positions of them and optimized deployment positions for near space vehicles can be obtained with iterative method. For controllable backbone nodes on earth surface, their deployment is executed within each groups which they belong and thus one particle represents position of one backbone node. During network initial stage, the deployment aims at maximizing the number of common nodes covered by backbone nodes and minimizing average hops from common nodes to backbone nodes. During network running stage, positions of backbone nodes are dynamically adjusted to best serve their groups. The dynamic deployment mechanism ensures backbone nodes are closest to the most heavily loaded nodes, or to the nodes that generate packets of the highest priority. Simulation results show that the proposed deployment scheme can provides better operation basis for upper layer protocols in terms of network performance metrics such as throughput, latency and control overhead.
2011, 33(4): 922-926.
doi: 10.3724/SP.J.1146.2010.00792
Abstract:
The vector and scalar potentials Greens functions of the Horizontal Electric Dipole (HED) radiation in sea water are derived. The Discrete Complex Image Method (DCIM) which has been used broadly in high frequency domain is introduced into the Extremely Low Frequency (ELF) electromagnetic wave propagation calculation in sea water. The integral path does not hold valid for the ELF situation and the reason can be ascribed to the change of the wave number at different frequency. The new integral path suitable for ELF situation is proposed. The sea environment is modeled as air-sea-sediment three-layer geometry. Generalized Pencil-Of- Function (GPOF) method is used to approximate the integral kernel with exponential summation. The DCIM is employed to obtain spatial domain closed-form Green's functions for the layered model. The electric filed is expressed based on the Mixed-Potential electric field Integral Equation. The results are compared with software FEKO and WHAM. It is safe to conclude that the two-level DCIM approach in this paper is effective and can exhibit better field decay rate at a further distance than WHAM.
The vector and scalar potentials Greens functions of the Horizontal Electric Dipole (HED) radiation in sea water are derived. The Discrete Complex Image Method (DCIM) which has been used broadly in high frequency domain is introduced into the Extremely Low Frequency (ELF) electromagnetic wave propagation calculation in sea water. The integral path does not hold valid for the ELF situation and the reason can be ascribed to the change of the wave number at different frequency. The new integral path suitable for ELF situation is proposed. The sea environment is modeled as air-sea-sediment three-layer geometry. Generalized Pencil-Of- Function (GPOF) method is used to approximate the integral kernel with exponential summation. The DCIM is employed to obtain spatial domain closed-form Green's functions for the layered model. The electric filed is expressed based on the Mixed-Potential electric field Integral Equation. The results are compared with software FEKO and WHAM. It is safe to conclude that the two-level DCIM approach in this paper is effective and can exhibit better field decay rate at a further distance than WHAM.
2011, 33(4): 927-931.
doi: 10.3724/SP.J.1146.2010.00685
Abstract:
Although the algorithms based on the Method of Characteristics (MoC) can ensure the transmission-line causality, there is no guarantee of passivity of the resulting macromodels. This paper proposes a new perturbation method for the passivity enforcement of MoC-based macromodels. The method generalizes the recently developed method that is introduced for lumped macromodels, which is based on quadratic programming with equality constraint. This associated optimization problem can be solved by the method of Lagrange multipliers. Numerical examples show that the presented method yields accurate passive macromodels in a limited simulation time.
Although the algorithms based on the Method of Characteristics (MoC) can ensure the transmission-line causality, there is no guarantee of passivity of the resulting macromodels. This paper proposes a new perturbation method for the passivity enforcement of MoC-based macromodels. The method generalizes the recently developed method that is introduced for lumped macromodels, which is based on quadratic programming with equality constraint. This associated optimization problem can be solved by the method of Lagrange multipliers. Numerical examples show that the presented method yields accurate passive macromodels in a limited simulation time.
2011, 33(4): 932-937.
doi: 10.3724/SP.J.1146.2010.00776
Abstract:
Mixed-Polarity conversion is one of important phases in logic synthesis of Reed-Muller (RM) circuits, which implements the conversions from Boolean logic Minterm expressions to RM logic Mixed-Polarity Reed-Muller (MPRM) expressions. In this paper, based on the research of decomposed rules of Ordered Kronecker Functional Decision Diagrams (OKFDDs) the relations between MPRM expressions and OKFDDs database are established. On this basis, according to the subscripts included relations between Minterm coefficients and MPRM coefficients and combining the description of multi-output logic functions, a novel Mixed-Polarity conversion algorithm directly from Minterm expressions to MPRM expressions is proposed. Finally, through several Benchmark tests, the results show the efficiency of the methods, which is significantly improved compared to other conversion algorithms.
Mixed-Polarity conversion is one of important phases in logic synthesis of Reed-Muller (RM) circuits, which implements the conversions from Boolean logic Minterm expressions to RM logic Mixed-Polarity Reed-Muller (MPRM) expressions. In this paper, based on the research of decomposed rules of Ordered Kronecker Functional Decision Diagrams (OKFDDs) the relations between MPRM expressions and OKFDDs database are established. On this basis, according to the subscripts included relations between Minterm coefficients and MPRM coefficients and combining the description of multi-output logic functions, a novel Mixed-Polarity conversion algorithm directly from Minterm expressions to MPRM expressions is proposed. Finally, through several Benchmark tests, the results show the efficiency of the methods, which is significantly improved compared to other conversion algorithms.
2011, 33(4): 938-944.
doi: 10.3724/SP.J.1146.2010.00714
Abstract:
This paper proposes a novel delay-independent communication interface used in multiprocessor System- on-Chip (SoC) and Network-on-Chip (NoC). Data can be transferred fast from asynchronous to synchronous through the interface gracefully, which is implemented by the circular FIFO handled under special operation protocol, and various asynchronous transfer protocols are supported. Meanwhile, the communication integrity and high throughput are maintained during transmission. Simulations are made based on SMIC 0.18 m CMOS technology. Results show that the delay is 792 ps with the average energy consumption of 4.87 pJ/request, which can satisfy the requirements of high speed low power, strong robustness and good reusability in Multiprocessor SoC and NoC.
This paper proposes a novel delay-independent communication interface used in multiprocessor System- on-Chip (SoC) and Network-on-Chip (NoC). Data can be transferred fast from asynchronous to synchronous through the interface gracefully, which is implemented by the circular FIFO handled under special operation protocol, and various asynchronous transfer protocols are supported. Meanwhile, the communication integrity and high throughput are maintained during transmission. Simulations are made based on SMIC 0.18 m CMOS technology. Results show that the delay is 792 ps with the average energy consumption of 4.87 pJ/request, which can satisfy the requirements of high speed low power, strong robustness and good reusability in Multiprocessor SoC and NoC.
2011, 33(4): 945-950.
doi: 10.3724/SP.J.1146.2010.00895
Abstract:
The power consumption, wiring overhead and crosstalk in deep sub-micron on-chip buses are the main facts restricting the bus throughput. An adaptive spatio-temporal bus coding scheme is proposed to reduce crosstalk induced delay and power consumption in the buses. Firstly, on-chip bus is partitioned into two sub-buses by spatio coding to reduce the Worst-Case-Crosstalk (WCC). Then, decisions are made respectively in the two sub-buses whether the original code and inverted code should incur WCC through an arbiter CCA (Crosstalk Class Arbiter). In the case of both original and inverted codes in any sub-bus incurring WCC, a shielding pattern is transmitted; otherwise the WCC-free and energy saving code is transmitted. The proposed scheme is evaluated using the SPEC benchmarks. The results show that the proposed scheme improves the throughput by 62.59% to 81.62% over the un-coded approach and reduces the power consumption by 14.63% to 54.67% compared to the other similar schemes while eliminating the WCC with only 7 wires overhead for a 32 bit bus. The scheme achieves a good enhancement in dynamic power, wiring overhead, and performance.
The power consumption, wiring overhead and crosstalk in deep sub-micron on-chip buses are the main facts restricting the bus throughput. An adaptive spatio-temporal bus coding scheme is proposed to reduce crosstalk induced delay and power consumption in the buses. Firstly, on-chip bus is partitioned into two sub-buses by spatio coding to reduce the Worst-Case-Crosstalk (WCC). Then, decisions are made respectively in the two sub-buses whether the original code and inverted code should incur WCC through an arbiter CCA (Crosstalk Class Arbiter). In the case of both original and inverted codes in any sub-bus incurring WCC, a shielding pattern is transmitted; otherwise the WCC-free and energy saving code is transmitted. The proposed scheme is evaluated using the SPEC benchmarks. The results show that the proposed scheme improves the throughput by 62.59% to 81.62% over the un-coded approach and reduces the power consumption by 14.63% to 54.67% compared to the other similar schemes while eliminating the WCC with only 7 wires overhead for a 32 bit bus. The scheme achieves a good enhancement in dynamic power, wiring overhead, and performance.
2011, 33(4): 951-955.
doi: 10.3724/SP.J.1146.2010.00789
Abstract:
In this paper, a Field Effect Transistor (FET) operating in close to cut off frequency is used to replace the attenuator and the phase-shifter in most TWT predistortion circuit, which can realize good control for the amount of gain expansion with little phase expansion variation. The experiments with practically designed circuit show that, in the frequency range of 8.38~8.58 GHz, about 6 dB gain expansion and 45 degrees phase shift can be compensated for the TWTA saturation response in input power dynamic range, and the improvement of about 15.4 dB Carrier to Intermodulation (C/IM) at TWTA Output Power Back Off (OPBO) of only -3 dB can be reached with the linearizer.
In this paper, a Field Effect Transistor (FET) operating in close to cut off frequency is used to replace the attenuator and the phase-shifter in most TWT predistortion circuit, which can realize good control for the amount of gain expansion with little phase expansion variation. The experiments with practically designed circuit show that, in the frequency range of 8.38~8.58 GHz, about 6 dB gain expansion and 45 degrees phase shift can be compensated for the TWTA saturation response in input power dynamic range, and the improvement of about 15.4 dB Carrier to Intermodulation (C/IM) at TWTA Output Power Back Off (OPBO) of only -3 dB can be reached with the linearizer.
2011, 33(4): 956-961.
doi: 10.3724/SP.J.1146.2010.00795
Abstract:
The theory of periodic permanent magnetic focusing system is studied. 2D second-order finite element method is used to simulate magnetic displacement of axisymmetric permanent magnetic models in cylindrical coordinates. First the convergence is compared between linear finite element method and second-order finite element method. Second-order finite element method has better convergence and faster convergence rate. Then the single magnetic ring model and the single period model are simulated. These results are compared with the results of Ansoft Maxwell 2D and UESTC-PPM. 2D second-order finite element method is more accurate.
The theory of periodic permanent magnetic focusing system is studied. 2D second-order finite element method is used to simulate magnetic displacement of axisymmetric permanent magnetic models in cylindrical coordinates. First the convergence is compared between linear finite element method and second-order finite element method. Second-order finite element method has better convergence and faster convergence rate. Then the single magnetic ring model and the single period model are simulated. These results are compared with the results of Ansoft Maxwell 2D and UESTC-PPM. 2D second-order finite element method is more accurate.
2011, 33(4): 962-966.
doi: 10.3724/SP.J.1146.2010.00787
Abstract:
Based on the image differences, Intrinsicface is proposed, which divides the face image into three parts, common facial differences, intrapersonal differences and individual differences, and shows desirable performance. But it does not consider the manifold structure and suffers from the singular problem, which is also called Small Sample Size (SSS) problem. To solve these problems, Null Space Locality Preserving Discriminant Intrinsicface (NSLPDI) is proposed, which makes full use of intrapersonal differences and individual differences and employs the idea of manifold learning so that the similarity in the intra-class is preserved while the separability of samples from different classes is enlarged by discriminant criterion. The optimal feature vectors are extracted from the null space of intrapersonal locality preserving difference scatter matrix, which avoids the singularity and the SSS problem is solved. Experiments on face recognition demonstrate the correctness and effectiveness of the proposed algorithm.
Based on the image differences, Intrinsicface is proposed, which divides the face image into three parts, common facial differences, intrapersonal differences and individual differences, and shows desirable performance. But it does not consider the manifold structure and suffers from the singular problem, which is also called Small Sample Size (SSS) problem. To solve these problems, Null Space Locality Preserving Discriminant Intrinsicface (NSLPDI) is proposed, which makes full use of intrapersonal differences and individual differences and employs the idea of manifold learning so that the similarity in the intra-class is preserved while the separability of samples from different classes is enlarged by discriminant criterion. The optimal feature vectors are extracted from the null space of intrapersonal locality preserving difference scatter matrix, which avoids the singularity and the SSS problem is solved. Experiments on face recognition demonstrate the correctness and effectiveness of the proposed algorithm.
2011, 33(4): 967-971.
doi: 10.3724/SP.J.1146.2010.00684
Abstract:
Based on the standard compressed sensing, the model-based Compressed Sensing (CS) uses the tree structure priors, and solves the optimal reconstruction problem with two existing tree structure approximation which are greedy tree approximation and optimal tree approximation. Through numerous statistics test of wavelet relationship, a new tree structure which is named reasonable tree structure is proposed, which is based on the relationship between neighbor coefficients, parent coefficients and children coefficients. What is more, combining with the new reasonable tree structure, an improvement is made for the iterative hard threshold reconstruction algorithm and model-based compressed sensing reconstruction algorithm. Comparing with the iterative hard threshold algorithm and model-based compressed sensing algorithm, the proposed algorithm can achieve higher image reconstruction performance.
Based on the standard compressed sensing, the model-based Compressed Sensing (CS) uses the tree structure priors, and solves the optimal reconstruction problem with two existing tree structure approximation which are greedy tree approximation and optimal tree approximation. Through numerous statistics test of wavelet relationship, a new tree structure which is named reasonable tree structure is proposed, which is based on the relationship between neighbor coefficients, parent coefficients and children coefficients. What is more, combining with the new reasonable tree structure, an improvement is made for the iterative hard threshold reconstruction algorithm and model-based compressed sensing reconstruction algorithm. Comparing with the iterative hard threshold algorithm and model-based compressed sensing algorithm, the proposed algorithm can achieve higher image reconstruction performance.
2011, 33(4): 972-976.
doi: 10.3724/SP.J.1146.2010.00783
Abstract:
Weighted Subspace Fitting (WSF) algorithm is a well-known excellent algorithm for DOA estimation with low SNR and few snapshots. However, this algorithm is totally impractical for its prohibitive computational burden incurred by multi-dimensional nonlinear search. In order to solve this problem, Ant Colony Optimization (ACO) is introduced to combine with the WSF algorithm and a new algorithm with lower computational burden called ACO-WSF is proposed. The proposed algorithm exploits Gaussian kernel probability density function in the sampling process. The global maximum of WSF spatial spectrum function can be reached after reasonable iterations. Simulation results illustrate that the proposed algorithm not only provides similar performance as WSF algorithm and better performance than MUSIC algorithm in the situation of low SNR and few snapshots, but also reduces computational complexity significantly.
Weighted Subspace Fitting (WSF) algorithm is a well-known excellent algorithm for DOA estimation with low SNR and few snapshots. However, this algorithm is totally impractical for its prohibitive computational burden incurred by multi-dimensional nonlinear search. In order to solve this problem, Ant Colony Optimization (ACO) is introduced to combine with the WSF algorithm and a new algorithm with lower computational burden called ACO-WSF is proposed. The proposed algorithm exploits Gaussian kernel probability density function in the sampling process. The global maximum of WSF spatial spectrum function can be reached after reasonable iterations. Simulation results illustrate that the proposed algorithm not only provides similar performance as WSF algorithm and better performance than MUSIC algorithm in the situation of low SNR and few snapshots, but also reduces computational complexity significantly.
2011, 33(4): 977-981.
doi: 10.3724/SP.J.1146.2010.00599
Abstract:
For multi-sensor distributed and quantitative estimation fusion problem of underwater target detection, a model of distributed and quantitative estimation fusion is established. The channel noise and its statistical property which is not fully known to fusion center is considered, The superiority of Expectation Maximization (EM) algorithm completely is used in parameter estimation problem when the observation data is missing. A new algorithm of distributed and quantitative estimation fusion is proposed based on EM algorithm. In this method, the unknown parameters of underwater acoustic channel noise and the quantization probability of local quantizer are modeled as the binary Gaussian mixture model parameters. Then, the invariance of the maximum likelihood estimation is used to get the result of the estimation fusion. Simulation results show that the estimation performance of the new algorithm is comparable to the methods which need ideal channel condition when the number of local sensors samples is larger than 5000 and the signal to noise ratio is higher than 6 dB. This new algorithm provides a theoretical basis for realizing the distributed and quantitative estimation fusion system of underwater target detection.
For multi-sensor distributed and quantitative estimation fusion problem of underwater target detection, a model of distributed and quantitative estimation fusion is established. The channel noise and its statistical property which is not fully known to fusion center is considered, The superiority of Expectation Maximization (EM) algorithm completely is used in parameter estimation problem when the observation data is missing. A new algorithm of distributed and quantitative estimation fusion is proposed based on EM algorithm. In this method, the unknown parameters of underwater acoustic channel noise and the quantization probability of local quantizer are modeled as the binary Gaussian mixture model parameters. Then, the invariance of the maximum likelihood estimation is used to get the result of the estimation fusion. Simulation results show that the estimation performance of the new algorithm is comparable to the methods which need ideal channel condition when the number of local sensors samples is larger than 5000 and the signal to noise ratio is higher than 6 dB. This new algorithm provides a theoretical basis for realizing the distributed and quantitative estimation fusion system of underwater target detection.
2011, 33(4): 982-986.
doi: 10.3724/SP.J.1146.2010.00691
Abstract:
With the development of wideband radar technology, classic narrowband Cramer-Rao (CR) bound can not be used as the criteria for the parameters estimation. Based on the application of Wideband Stepped Frequency Signal (WSFS), the wideband CR bound for WSFS parameters estimation is firstly derived in this paper, and the complete expression of Fisher information matrix is provided. Then an insight into theoretical performance of the wideband CR bound is given, from which it can be known that the estimated performance of stepped frequency signal is not only determined by the signal bandwidth and signal-to-noise ratio, but also affected by the relative bandwidth. Finally, it is proved that the narrowband CR bound is just a special case of wideband CR bound. All factors of wideband model are fully considered in the wideband CR bound, which make it more comprehensive in the theoretical analysis.
With the development of wideband radar technology, classic narrowband Cramer-Rao (CR) bound can not be used as the criteria for the parameters estimation. Based on the application of Wideband Stepped Frequency Signal (WSFS), the wideband CR bound for WSFS parameters estimation is firstly derived in this paper, and the complete expression of Fisher information matrix is provided. Then an insight into theoretical performance of the wideband CR bound is given, from which it can be known that the estimated performance of stepped frequency signal is not only determined by the signal bandwidth and signal-to-noise ratio, but also affected by the relative bandwidth. Finally, it is proved that the narrowband CR bound is just a special case of wideband CR bound. All factors of wideband model are fully considered in the wideband CR bound, which make it more comprehensive in the theoretical analysis.
2011, 33(4): 987-991.
doi: 10.3724/SP.J.1146.2010.00794
Abstract:
As the application of SATCOM On-The-Move (SOTM) is restricted by the height of antenna, the planar antenna with multiple subarrays is adopted to reduce the height of antenna. However, the excursion of beam pointing versus frequency is deteriorated because of its dispersed aperture. In order to reduce the excursion of beam pointing, the feeder system using fixed delay line and phase shifter is applied in this paper, and the switching-beam technique based on multi-set fixed delay lines is proposed. The proposed method divides the antenna scanning range into several small ones, and uses different fixed delay lines in each range. Simulation results show that the method can reduce the excursion of beam pointing effectively, and the requirement of beam pointing accuracy for SOTM can be satisfied.
As the application of SATCOM On-The-Move (SOTM) is restricted by the height of antenna, the planar antenna with multiple subarrays is adopted to reduce the height of antenna. However, the excursion of beam pointing versus frequency is deteriorated because of its dispersed aperture. In order to reduce the excursion of beam pointing, the feeder system using fixed delay line and phase shifter is applied in this paper, and the switching-beam technique based on multi-set fixed delay lines is proposed. The proposed method divides the antenna scanning range into several small ones, and uses different fixed delay lines in each range. Simulation results show that the method can reduce the excursion of beam pointing effectively, and the requirement of beam pointing accuracy for SOTM can be satisfied.
2011, 33(4): 992-996.
doi: 10.3724/SP.J.1146.2010.00756
Abstract:
The Asymptotic Waveform Evaluation (AWE) technique combined with the Method of Moments (MoM) is often applied to accelerate the calculation of the wide-band Radar Cross Section (RCS) of the target. However, when the object is electrically large, the method is still quite time-consuming. Hence, in this paper, a new type of Krylov subspace iterative methods based on flexible inner-outer iterative technique is adopted to solve the large dense matrix equation arising from the discretization of the Electric Field Integral Equation (EFIE). Near-field matrix preconditioning technique combined with the dual threshold incomplete LU factorization preconditioning technique will be applied to the reduction of the iterative number of FBICGSTAB. The numerical results show that the computational efficiency of the wide-band RCS for electrically large objects is improved significantly by using the method presented in this paper without sacrificing much accuracy.
The Asymptotic Waveform Evaluation (AWE) technique combined with the Method of Moments (MoM) is often applied to accelerate the calculation of the wide-band Radar Cross Section (RCS) of the target. However, when the object is electrically large, the method is still quite time-consuming. Hence, in this paper, a new type of Krylov subspace iterative methods based on flexible inner-outer iterative technique is adopted to solve the large dense matrix equation arising from the discretization of the Electric Field Integral Equation (EFIE). Near-field matrix preconditioning technique combined with the dual threshold incomplete LU factorization preconditioning technique will be applied to the reduction of the iterative number of FBICGSTAB. The numerical results show that the computational efficiency of the wide-band RCS for electrically large objects is improved significantly by using the method presented in this paper without sacrificing much accuracy.
2011, 33(4): 997-1001.
doi: 10.3724/SP.J.1146.2010.00471
Abstract:
When the feedback structure based on COSTAS loop is used in carrier synchronization of OQPSK signal with large frequency offset, the locking time is so long that feedback structure poorly fits for burst signal. To resolve this problem, this paper designs an improved algorithm based on the power spectrum of OQPSK signal and the feedback structure. Simulation results show that the performance of the improved algorithm is better than the traditional COSTAS loop, and the improved algorithm has a short locking time and low bit error rate which almost attains Shannon capacity under that power-to-noise ratio is higher than 6 dB.
When the feedback structure based on COSTAS loop is used in carrier synchronization of OQPSK signal with large frequency offset, the locking time is so long that feedback structure poorly fits for burst signal. To resolve this problem, this paper designs an improved algorithm based on the power spectrum of OQPSK signal and the feedback structure. Simulation results show that the performance of the improved algorithm is better than the traditional COSTAS loop, and the improved algorithm has a short locking time and low bit error rate which almost attains Shannon capacity under that power-to-noise ratio is higher than 6 dB.
2011, 33(4): 1002-1006.
doi: 10.3724/SP.J.1146.2010.00721
Abstract:
The communication simulation model of HF multitone parallel transmission is established，and the anti-jamming performance of narrowband and broadband is analyzed based on the model. A new method of subcarriers skipping anti-jamming for HF Tactical Data Links (TDLs) is presented aiming at narrowband multitone jamming, and simulations are made under the multiple conditions of skimming in or outside the channel bandwidth based on the communication simulation model. The simulation results show that the HF TDLs communication based on multitone parallel transmission can avoid narrowband jamming effectively such as multitone jamming.
The communication simulation model of HF multitone parallel transmission is established，and the anti-jamming performance of narrowband and broadband is analyzed based on the model. A new method of subcarriers skipping anti-jamming for HF Tactical Data Links (TDLs) is presented aiming at narrowband multitone jamming, and simulations are made under the multiple conditions of skimming in or outside the channel bandwidth based on the communication simulation model. The simulation results show that the HF TDLs communication based on multitone parallel transmission can avoid narrowband jamming effectively such as multitone jamming.
2011, 33(4): 1007-1011.
doi: 10.3724/SP.J.1146.2010.00758
Abstract:
In this paper, a novel fusion based coding scheme of multi-view image is proposed. At the encoder, different sources of images are decomposed into IMFs (Intrinsic Mode Factions) and residue using CEMD (Complex Empirical Mode Decomposition). The IMFs and residue are fused respectively, and the fusion results are compressed and encoded based on EMD. At the decoder, the source images can be reconstructed by disassembling the fusion image. Experimental results show, without reducing quality, the proposed scheme can provide better compression ratio than the previous ones, which open doors for future applications.
In this paper, a novel fusion based coding scheme of multi-view image is proposed. At the encoder, different sources of images are decomposed into IMFs (Intrinsic Mode Factions) and residue using CEMD (Complex Empirical Mode Decomposition). The IMFs and residue are fused respectively, and the fusion results are compressed and encoded based on EMD. At the decoder, the source images can be reconstructed by disassembling the fusion image. Experimental results show, without reducing quality, the proposed scheme can provide better compression ratio than the previous ones, which open doors for future applications.
2011, 33(4): 1012-1016.
doi: 10.3724/SP.J.1146.2010.00712
Abstract:
Having an effective consistency maintenance mechanism is one of keys to guarantee P2P systems running in good condition. In unstructured P2P systems, message transmission is mainly by neighbors so that many redundant messages are produced. Moreover, time cost is increased due to logical-location and underlying physical-location of peer mismatching. To reduce consistency maintenance cost, this paper presents a replication consistency maintenance mechanism based on physical-location and Bloom-filter. According to their physical-location using community partition in complex networks, it partitions peers into clusters, where physical-location among peers is closer and updating is done. And it uses double-deck Bloom-filter and redundancy secondary neighbor to improve transmitting method of updating-message. The experimental results show this new mechanism can reduce updating-message redundancy and time cost of consistency maintenance.
Having an effective consistency maintenance mechanism is one of keys to guarantee P2P systems running in good condition. In unstructured P2P systems, message transmission is mainly by neighbors so that many redundant messages are produced. Moreover, time cost is increased due to logical-location and underlying physical-location of peer mismatching. To reduce consistency maintenance cost, this paper presents a replication consistency maintenance mechanism based on physical-location and Bloom-filter. According to their physical-location using community partition in complex networks, it partitions peers into clusters, where physical-location among peers is closer and updating is done. And it uses double-deck Bloom-filter and redundancy secondary neighbor to improve transmitting method of updating-message. The experimental results show this new mechanism can reduce updating-message redundancy and time cost of consistency maintenance.
