Email alert
2013 Vol. 35, No. 10
Display Method:
2013, 35(10): 2287-2293.
doi: 10.3724/SP.J.1146.2013.00131
Abstract:
In the past few years, as a type of image authentication technique without relying on pre-registration or pre-embedded information, the passive blind image forensics has become a hot issue in the field of information security techniques. In this paper, a novel algorithm for detecting smoothing filtering in digital images is proposed based on the frequency residual. The suspected image is re-filtered with a Gaussian low-pass filter, and the difference between the initial image and the re-filtered image in Fourier domain is called the frequency residual. Then, the frequency residual is projected into the Radon space with an adaptation of Radon transform. The obtained data is modeled as Fourier series and the model parameters are adopted as features for filtering detection. The experimental results show that the proposed algorithm can not only detect three typical smoothing spatial filters, including Gaussian filter, average filter, and median filter, but also can predict parameters of these filters to complement the existing state-of-the-art methods.
In the past few years, as a type of image authentication technique without relying on pre-registration or pre-embedded information, the passive blind image forensics has become a hot issue in the field of information security techniques. In this paper, a novel algorithm for detecting smoothing filtering in digital images is proposed based on the frequency residual. The suspected image is re-filtered with a Gaussian low-pass filter, and the difference between the initial image and the re-filtered image in Fourier domain is called the frequency residual. Then, the frequency residual is projected into the Radon space with an adaptation of Radon transform. The obtained data is modeled as Fourier series and the model parameters are adopted as features for filtering detection. The experimental results show that the proposed algorithm can not only detect three typical smoothing spatial filters, including Gaussian filter, average filter, and median filter, but also can predict parameters of these filters to complement the existing state-of-the-art methods.
2013, 35(10): 2294-2300.
doi: 10.3724/SP.J.1146.2012.01719
Abstract:
The necessary condition of the optimal pixel expansion is given and proved in (k, n) XOR-based Visual Cryptography Scheme (XVCS). A new method is designed for constructing (k, n)-XVCS by use of basis matrices. The method is proved to be suitable for (k, n)-XVCS with 2 kn. Based on the above results, the secret sharing and recovering algorithms are proposed. The experimental results show that the pixel expansion can be decreased efficiently. Furthermore, the secret images can be recovered perfectly.
The necessary condition of the optimal pixel expansion is given and proved in (k, n) XOR-based Visual Cryptography Scheme (XVCS). A new method is designed for constructing (k, n)-XVCS by use of basis matrices. The method is proved to be suitable for (k, n)-XVCS with 2 kn. Based on the above results, the secret sharing and recovering algorithms are proposed. The experimental results show that the pixel expansion can be decreased efficiently. Furthermore, the secret images can be recovered perfectly.
2013, 35(10): 2301-2306.
doi: 10.3724/SP.J.1146.2012.01615
Abstract:
ARIA cipher is a new block cipher proposed by some South Korean experts in 2003. The design principle of ARIA is similar to the AES. ARIA is established as a Korean Standard block cipher algorithm by Korean Agency for Technology and Standards. In this paper, a new impossible differential attack on 7 rounds of the ARIA cipher is presented. By this attack, 7-round ARIA-192 is breakable with a data complexity of about2176.2 encryptions,while the previous best impossible differential attack on ARIA broke on 7-round ARIA-256. Then some characters on diffusion layer are used to reduce the complexity of the ARIA-256 to 2192.2.
ARIA cipher is a new block cipher proposed by some South Korean experts in 2003. The design principle of ARIA is similar to the AES. ARIA is established as a Korean Standard block cipher algorithm by Korean Agency for Technology and Standards. In this paper, a new impossible differential attack on 7 rounds of the ARIA cipher is presented. By this attack, 7-round ARIA-192 is breakable with a data complexity of about2176.2 encryptions,while the previous best impossible differential attack on ARIA broke on 7-round ARIA-256. Then some characters on diffusion layer are used to reduce the complexity of the ARIA-256 to 2192.2.
2013, 35(10): 2307-2313.
doi: 10.3724/SP.J.1146.2013.00114
Abstract:
RoQ (Reduction of Quality) attack is more stealthy and changeable than traditional DoS (Denial of Service) attack, which makes detection of RoQ extremely difficult. In order to improve detection accuracy and locate attack sources in time, this paper turns modeling attack flow extraction into a process of blind sources separation. A method is proposed based on fast ICA (Independent Component Analysis) to detach RoQ flow from several observation network devices and terminals. Then, some features parameters that represent attack flow are extracted. After that, a system of collaborative detection system is designed on the basis of SVM (Support Vector Machine), using marked attack and no-attack samples to train the SVM classifier in order to detect RoQ attack finally. Simulation results illustrate that this method can detect IP spoofed RoQ attack as well as locate the attacker, accuracy of which reaches up to 90%. Moreover, choosing appropriate ICA parameters will improve results to some extent.
RoQ (Reduction of Quality) attack is more stealthy and changeable than traditional DoS (Denial of Service) attack, which makes detection of RoQ extremely difficult. In order to improve detection accuracy and locate attack sources in time, this paper turns modeling attack flow extraction into a process of blind sources separation. A method is proposed based on fast ICA (Independent Component Analysis) to detach RoQ flow from several observation network devices and terminals. Then, some features parameters that represent attack flow are extracted. After that, a system of collaborative detection system is designed on the basis of SVM (Support Vector Machine), using marked attack and no-attack samples to train the SVM classifier in order to detect RoQ attack finally. Simulation results illustrate that this method can detect IP spoofed RoQ attack as well as locate the attacker, accuracy of which reaches up to 90%. Moreover, choosing appropriate ICA parameters will improve results to some extent.
2013, 35(10): 2314-2320.
doi: 10.3724/SP.J.1146.2013.00545
Abstract:
Web service selection is a critical procedure for performance-enhancing in composite service. To find the best services from the candidate services, both the Quality of Service (QoS) requirements and functional requirements should be considered. However, most Web service selection methods are based on the assumption of the independence among candidate services, and ignore the dependency relationship and compatible relationship among candidate services. In practice, composite services emphasize the coordination among these component services. The function of a candidate service in a composite service usually depends on the other optional service. To solve this problem, a multi-constraint service selection method is proposed based on local approximate filter. This method filters out part of the unsatisfied constrain services based on local approximate filter, and estimates the local fitness of each of the rest candidate services, then defines a suitable particle swarm algorithm to search the optimal solutions in the light of the calculated local fitness. Experimental results demonstrate the effectiveness of this method.
Web service selection is a critical procedure for performance-enhancing in composite service. To find the best services from the candidate services, both the Quality of Service (QoS) requirements and functional requirements should be considered. However, most Web service selection methods are based on the assumption of the independence among candidate services, and ignore the dependency relationship and compatible relationship among candidate services. In practice, composite services emphasize the coordination among these component services. The function of a candidate service in a composite service usually depends on the other optional service. To solve this problem, a multi-constraint service selection method is proposed based on local approximate filter. This method filters out part of the unsatisfied constrain services based on local approximate filter, and estimates the local fitness of each of the rest candidate services, then defines a suitable particle swarm algorithm to search the optimal solutions in the light of the calculated local fitness. Experimental results demonstrate the effectiveness of this method.
2013, 35(10): 2321-2327.
doi: 10.3724/SP.J.1146.2012.01715
Abstract:
In a wireless acoustic sensor network where microphones are distributedly scattered, the microphone subset selection is usually based on the quality of the received signals on the microphones. However, it is difficult to evaluate it without knowing about the nodes and the sound sources position, the source signal and the noise signals. In this paper, a novel algorithm is proposed depending only on the received data for assessing microphone utility blindly. The algorithm exploits the relationship between the SNR and the signals higher-order statistical information (kurtosis) in frequency domain and the microphone utility is a weighted sum of the kurtosis of each frequency bin. The experimental results show that the proposed algorithm can effectively evaluate the quality of the received signals and get the comparable performance with the true SNR.
In a wireless acoustic sensor network where microphones are distributedly scattered, the microphone subset selection is usually based on the quality of the received signals on the microphones. However, it is difficult to evaluate it without knowing about the nodes and the sound sources position, the source signal and the noise signals. In this paper, a novel algorithm is proposed depending only on the received data for assessing microphone utility blindly. The algorithm exploits the relationship between the SNR and the signals higher-order statistical information (kurtosis) in frequency domain and the microphone utility is a weighted sum of the kurtosis of each frequency bin. The experimental results show that the proposed algorithm can effectively evaluate the quality of the received signals and get the comparable performance with the true SNR.
2013, 35(10): 2328-2334.
doi: 10.3724/SP.J.1146.2013.00056
Abstract:
Information of sensor percept are send to interested target node through sensor network. The multicast technology in traditional networks are usually more energy consumption, worse real-time. It is not fitted in sensor network. For nodes is unknown of network topology, multicast routing issue is converted to the optimal multicast path issue firstly, and it is solved through the heuristic algorithm. The multicast routing tree is optimized through cutting merger strategy with greedy, the optimal path is got in the entire network. Finally, combined with the node area concentration and wireless multicast characteristics, a routing algorithm, named DCast, is proposed. Finally, simulation results demonstrate the advantages of energy consumption and delay compared with other classic multicast routing algorithm such as uCast, SenCast.
Information of sensor percept are send to interested target node through sensor network. The multicast technology in traditional networks are usually more energy consumption, worse real-time. It is not fitted in sensor network. For nodes is unknown of network topology, multicast routing issue is converted to the optimal multicast path issue firstly, and it is solved through the heuristic algorithm. The multicast routing tree is optimized through cutting merger strategy with greedy, the optimal path is got in the entire network. Finally, combined with the node area concentration and wireless multicast characteristics, a routing algorithm, named DCast, is proposed. Finally, simulation results demonstrate the advantages of energy consumption and delay compared with other classic multicast routing algorithm such as uCast, SenCast.
2013, 35(10): 2335-2340.
doi: 10.3724/SP.J.1146.2012.01676
Abstract:
In the context of social network becomes more and more complicated and huge, it is difficult to improve the accuracy and performance of existing community detection algorithms only relying on the network topological features. Based on Markov random walk theory, this paper proposes a method of edge weighted pre-processing for optimizing community detection, models community structures how to influence on the complex network behaviors. According to the situation of multiple random walk traverses on the network links, the network edges weight is reset, and makes it as the network topology effective supplementary information to promote the network community structure defuzzification, thus the performance of the existing algorithms is improved for community detection. For a set of typical benchmark computer-generated networks and real-world network data sets, the experimental results show that the pre-processing method can effectively improve the accuracy and efficiency of some existing community detection algorithms.
In the context of social network becomes more and more complicated and huge, it is difficult to improve the accuracy and performance of existing community detection algorithms only relying on the network topological features. Based on Markov random walk theory, this paper proposes a method of edge weighted pre-processing for optimizing community detection, models community structures how to influence on the complex network behaviors. According to the situation of multiple random walk traverses on the network links, the network edges weight is reset, and makes it as the network topology effective supplementary information to promote the network community structure defuzzification, thus the performance of the existing algorithms is improved for community detection. For a set of typical benchmark computer-generated networks and real-world network data sets, the experimental results show that the pre-processing method can effectively improve the accuracy and efficiency of some existing community detection algorithms.
2013, 35(10): 2341-2346.
doi: 10.3724/SP.J.1146.2013.00153
Abstract:
With increase of storage devices, the data placements based on toleration single or double failures can not meet the requirement of the reliability in the distributed storage systems. On the basis of the Row Diagonal Parity (RDP) code for double toleration failures, a new class of array codes for triple storage failures is presented which is called Extending Row Diagonal Parity (E-RDP) code. The E-RDP code has the Maximum Distance Separable (MDS) property, and it is optimal in redundancy rate and erasure correcting capability among triple erasure-correcting codes. The procedures of encoding and decoding are depicted by geometrical lines of different slope, then a fast decoding algorithm is given and it is more easily implemented by software and hardware. The theoretical analysis shows that the comprehensive properties of the E-RDP code such as encoding and decoding efficiency, small writes and balance performance, are better than other popular MDS codes, thus the E-RDP code is practically meaningful for storage systems.
With increase of storage devices, the data placements based on toleration single or double failures can not meet the requirement of the reliability in the distributed storage systems. On the basis of the Row Diagonal Parity (RDP) code for double toleration failures, a new class of array codes for triple storage failures is presented which is called Extending Row Diagonal Parity (E-RDP) code. The E-RDP code has the Maximum Distance Separable (MDS) property, and it is optimal in redundancy rate and erasure correcting capability among triple erasure-correcting codes. The procedures of encoding and decoding are depicted by geometrical lines of different slope, then a fast decoding algorithm is given and it is more easily implemented by software and hardware. The theoretical analysis shows that the comprehensive properties of the E-RDP code such as encoding and decoding efficiency, small writes and balance performance, are better than other popular MDS codes, thus the E-RDP code is practically meaningful for storage systems.
2013, 35(10): 2347-2353.
doi: 10.3724/SP.J.1146.2012.01725
Abstract:
This paper proposes a proxy-based proactive caching approach for enhancing consumer mobility in Content Centric Networks (CCN). The approach requests proactively and caches the contents that a user does not receive but is requested before a handover process. Compared with other approaches for consumer mobility, the analytical and simulation results indicate that the proposed approach has shorter handover delay and less average content retrieval time. In addition, the proposed approach costs less in retrieving content for mobile users. This approach can also support efficient content sharing in mobile CCN environments and alleviate negative effects of consumer mobility on real-time services.
This paper proposes a proxy-based proactive caching approach for enhancing consumer mobility in Content Centric Networks (CCN). The approach requests proactively and caches the contents that a user does not receive but is requested before a handover process. Compared with other approaches for consumer mobility, the analytical and simulation results indicate that the proposed approach has shorter handover delay and less average content retrieval time. In addition, the proposed approach costs less in retrieving content for mobile users. This approach can also support efficient content sharing in mobile CCN environments and alleviate negative effects of consumer mobility on real-time services.
2013, 35(10): 2354-2358.
doi: 10.3724/SP.J.1146.2013.00130
Abstract:
This paper applies the random geometry theory to Decode-and-Forward (DF) relay-enhanced Orthogonal Frequency Division Multiplexing (OFDM) system and derives a new system capacity model finally. There are some assumptions that multiple users distribute uniformly in a ring region around Base Station (BS) and frequency is reused existing among users in access zone. By using the upper bound of interfere integral value to alternate the actual interfere value when considering the co-channel interference, the closed expression of system capacity is obtained under the power constraints of BS and relay nodes. Simulation results show the validity of the system capacity model.
This paper applies the random geometry theory to Decode-and-Forward (DF) relay-enhanced Orthogonal Frequency Division Multiplexing (OFDM) system and derives a new system capacity model finally. There are some assumptions that multiple users distribute uniformly in a ring region around Base Station (BS) and frequency is reused existing among users in access zone. By using the upper bound of interfere integral value to alternate the actual interfere value when considering the co-channel interference, the closed expression of system capacity is obtained under the power constraints of BS and relay nodes. Simulation results show the validity of the system capacity model.
2013, 35(10): 2359-2364.
doi: 10.3724/SP.J.1146.2013.00042
Abstract:
Since the underwater acoustic channel suffers often severe frequency-dependent attenuation, low speed of wave propagation and excessive multipath delay spread, the implementation of spectrum detection in Cognitive Underwater Acoustic Communication (CUAC) becomes very difficult. Beside, there is no fusion center in Ad hoc underwater acoustic communication networks. Therefore, the centralized spectrum detection methods in CUAC are not available. Similar to Cognitive Radio (CR), since the spectrum utility in CUAC is also low, the spectrum is sparse. Based on compressed sensing and considering the specificity of underwater acoustic, compressed spectrum detection algorithm for cognitive radio is improved, and then two distributed cooperative spectrum detection methods, which are suitable for CUAC, are proposed for different scenarios (with and without channel state information). By strengthening among secondary users, the proposed algorithms obtain spatial diversity gains and exploit joint sparse structure to improve the performance of spectrum detection. Via distributed computation and localized optimization, the new schemes entail low computation and power overhead per cognitive users. Simulation results corroborate the effectiveness of the proposed methods in detecting the spectrum holes in underwater acoustic environment.
Since the underwater acoustic channel suffers often severe frequency-dependent attenuation, low speed of wave propagation and excessive multipath delay spread, the implementation of spectrum detection in Cognitive Underwater Acoustic Communication (CUAC) becomes very difficult. Beside, there is no fusion center in Ad hoc underwater acoustic communication networks. Therefore, the centralized spectrum detection methods in CUAC are not available. Similar to Cognitive Radio (CR), since the spectrum utility in CUAC is also low, the spectrum is sparse. Based on compressed sensing and considering the specificity of underwater acoustic, compressed spectrum detection algorithm for cognitive radio is improved, and then two distributed cooperative spectrum detection methods, which are suitable for CUAC, are proposed for different scenarios (with and without channel state information). By strengthening among secondary users, the proposed algorithms obtain spatial diversity gains and exploit joint sparse structure to improve the performance of spectrum detection. Via distributed computation and localized optimization, the new schemes entail low computation and power overhead per cognitive users. Simulation results corroborate the effectiveness of the proposed methods in detecting the spectrum holes in underwater acoustic environment.
2013, 35(10): 2365-2370.
doi: 10.3724/SP.J.1146.2013.00028
Abstract:
In order to reduce the computational complexity, a fast inter-frame Prediction Unit (PU) mode decision algorithm is proposed for High Efficiency Video Coding (HEVC) based on the strong temporal correlation in video sequences. The correlation of PU modes of Coding Unit (CU) between two adjacent frames is analyzed. Considering the probable objects motion in videos, the correlation of PU modes between the surrounding CUs of the corresponding located CUs in the previous frame and the current CU is also analyzed. Based on the analysis, the redundant PU modes are skipped for the current CU to reduce the coding computational complexity. The expriment results show that the proposed method save 31.30% of coding time compared with the current fast inter perdiction algorithm for HEVC under the condition of minimal lose of coding effencicy and PSNR.
In order to reduce the computational complexity, a fast inter-frame Prediction Unit (PU) mode decision algorithm is proposed for High Efficiency Video Coding (HEVC) based on the strong temporal correlation in video sequences. The correlation of PU modes of Coding Unit (CU) between two adjacent frames is analyzed. Considering the probable objects motion in videos, the correlation of PU modes between the surrounding CUs of the corresponding located CUs in the previous frame and the current CU is also analyzed. Based on the analysis, the redundant PU modes are skipped for the current CU to reduce the coding computational complexity. The expriment results show that the proposed method save 31.30% of coding time compared with the current fast inter perdiction algorithm for HEVC under the condition of minimal lose of coding effencicy and PSNR.
2013, 35(10): 2371-2377.
doi: 10.3724/SP.J.1146.2013.00022
Abstract:
An adaptive wavelet packet image compressed sensing is proposed, in which the wavelet packet transform is used to decompose the image. After the image is decomposed, the properties of each packet wavelet block are analyzed with the introduction of mathematical expectation and information entropy. According to the characteristic of each packet wavelet block, the signals are classified to four types of signal, that is the low frequency signal, no value signal, special processing signal and compressed sensing processing signal adaptively. Then the corresponding methods are designed to deal with different types of signal, which can adapt to the different characteristic of images. In this method, the quality of compressed sensing is improved, which is because sampling numbers can be adaptively selected according to different images and packet wavelet blocks. Experimental results show that, when the sampling number is the same, the proposed algorithm can not only greatly improve the reconstruction quality of image, but also reduce the computational complexity and required memory.
An adaptive wavelet packet image compressed sensing is proposed, in which the wavelet packet transform is used to decompose the image. After the image is decomposed, the properties of each packet wavelet block are analyzed with the introduction of mathematical expectation and information entropy. According to the characteristic of each packet wavelet block, the signals are classified to four types of signal, that is the low frequency signal, no value signal, special processing signal and compressed sensing processing signal adaptively. Then the corresponding methods are designed to deal with different types of signal, which can adapt to the different characteristic of images. In this method, the quality of compressed sensing is improved, which is because sampling numbers can be adaptively selected according to different images and packet wavelet blocks. Experimental results show that, when the sampling number is the same, the proposed algorithm can not only greatly improve the reconstruction quality of image, but also reduce the computational complexity and required memory.
2013, 35(10): 2378-2383.
doi: 10.3724/SP.J.1146.2013.00195
Abstract:
Because of the multi-colinearity of the ill-conditioned mixing signals, it is difficult to solve the issue of underdetermined blind sources separation for ill-conditioned mixing signals in noisy environment by Sparse Component Analysis (SCA). The model of the problem is built and the limitation of clustering methods to solve the problem is analyzed in this paper. Then a robust underdetermined blind sources separation algorithm based on SCA and Nonorthogonal Joint Diagonalization (NJD) is presented. NJD has the property that the mixing matrix is not necessarily unitary, which is used to solve the above problem in the novel algorithm. Simulation experiments show that the algorithm can improve the performance in separation performance, noise robust and ill-conditioned mixing robust compared with Cluster Guide Particle Swarm Optimization (CGPSO) algorithm.
Because of the multi-colinearity of the ill-conditioned mixing signals, it is difficult to solve the issue of underdetermined blind sources separation for ill-conditioned mixing signals in noisy environment by Sparse Component Analysis (SCA). The model of the problem is built and the limitation of clustering methods to solve the problem is analyzed in this paper. Then a robust underdetermined blind sources separation algorithm based on SCA and Nonorthogonal Joint Diagonalization (NJD) is presented. NJD has the property that the mixing matrix is not necessarily unitary, which is used to solve the above problem in the novel algorithm. Simulation experiments show that the algorithm can improve the performance in separation performance, noise robust and ill-conditioned mixing robust compared with Cluster Guide Particle Swarm Optimization (CGPSO) algorithm.
2013, 35(10): 2384-2390.
doi: 10.3724/SP.J.1146.2013.00417
Abstract:
A novel method is proposed for segmentation of moving object using motion feature of edge in infrared videos. At first, a new index Motion Saliency of Edge (MSE) is defined. MSE can reflect the motion feature of edge points of an image based on the spatial-temporal characteristic. The higher the MSE of an edge point is, the more likely it belongs to a moving object. The edge point with a high MSE value is extracted by using Otsus thresholding. The results obtained by Otsu are updated by using historical data. The edge points which are extracted and updated can be considered to be the seed of moving objects. At last, the segmented masks of moving objects are grown from the seeds by using a region growing method of layer-by-layer. The proposed method is successfully tested over three infrared image sequences and compared with the other two methods. The experiment results demonstrate that the proposed method has better performance of moving object segmentation with less effect of object-background misclassification in infrared videos.
A novel method is proposed for segmentation of moving object using motion feature of edge in infrared videos. At first, a new index Motion Saliency of Edge (MSE) is defined. MSE can reflect the motion feature of edge points of an image based on the spatial-temporal characteristic. The higher the MSE of an edge point is, the more likely it belongs to a moving object. The edge point with a high MSE value is extracted by using Otsus thresholding. The results obtained by Otsu are updated by using historical data. The edge points which are extracted and updated can be considered to be the seed of moving objects. At last, the segmented masks of moving objects are grown from the seeds by using a region growing method of layer-by-layer. The proposed method is successfully tested over three infrared image sequences and compared with the other two methods. The experiment results demonstrate that the proposed method has better performance of moving object segmentation with less effect of object-background misclassification in infrared videos.
2013, 35(10): 2391-2396.
doi: 10.3724/SP.J.1146.2012.01743
Abstract:
Recommendations from Collaborative Filtering (CF) recommender algorithms have low timeliness. To solve the problem, an information aging-based collaborative filtering algorithm is proposed by combining information age method. The algorithm builds a model to evaluate the timeliness of items based on users hit records to predict the probabilities of the items being clicked at the present time. To consider comprehensively users interests and the timeliness of items, the model and item-based collaborative filtering recommender algorithm are combined to find the nearest neighbor collection. Experimental results show that comparing with traditional collaborative filtering recommender algorithm the proposed algorithm can improve the timeliness of recommendations.
Recommendations from Collaborative Filtering (CF) recommender algorithms have low timeliness. To solve the problem, an information aging-based collaborative filtering algorithm is proposed by combining information age method. The algorithm builds a model to evaluate the timeliness of items based on users hit records to predict the probabilities of the items being clicked at the present time. To consider comprehensively users interests and the timeliness of items, the model and item-based collaborative filtering recommender algorithm are combined to find the nearest neighbor collection. Experimental results show that comparing with traditional collaborative filtering recommender algorithm the proposed algorithm can improve the timeliness of recommendations.
2013, 35(10): 2397-2402.
doi: 10.3724/SP.J.1146.2013.00102
Abstract:
A trajectory mapping method, called MagCom, is proposed based on the geomagnetic perception of indoor mobile devices. MagCom gets the initial position by means of Wi-Fi localization technology, and then calculates the instantaneous position of users continuously leveraging data of the phone's built-in inertial sensors in real time by the floors plan. The magnetic field data of the pedestrians location are captured to calibrate the mapping trajectory of mobile devices with the least squares algorithm. Furthermore, the unit vector and the norm vector are applied to reduce outliers in the searching algorithm. The experimental results show that the proposed method can not only improve the trajectory mapping accuracy but reduce the computational complexity than the state of arts.
A trajectory mapping method, called MagCom, is proposed based on the geomagnetic perception of indoor mobile devices. MagCom gets the initial position by means of Wi-Fi localization technology, and then calculates the instantaneous position of users continuously leveraging data of the phone's built-in inertial sensors in real time by the floors plan. The magnetic field data of the pedestrians location are captured to calibrate the mapping trajectory of mobile devices with the least squares algorithm. Furthermore, the unit vector and the norm vector are applied to reduce outliers in the searching algorithm. The experimental results show that the proposed method can not only improve the trajectory mapping accuracy but reduce the computational complexity than the state of arts.
2013, 35(10): 2403-2410.
doi: 10.3724/SP.J.1146.2012.01569
Abstract:
Facial expression recognition is a popular and difficult research field in human-computer interaction. In order to remove effectively the differences in expression feature caused by individual differences, this paper firstly presents the feature point distance ratio coefficient based on feature point vector, and then gives the concept of texture deformation energy parameters. Finally, merges previously mentioned two parts to form a new expression feature for facial expression recognition. The proposed method is tested in the Cohn-Kanade database and the BHU facial expression database, and the experimental results show the recognition rates of the proposed method comparing with the existing ones increased by 4.5% and 3.9%.
Facial expression recognition is a popular and difficult research field in human-computer interaction. In order to remove effectively the differences in expression feature caused by individual differences, this paper firstly presents the feature point distance ratio coefficient based on feature point vector, and then gives the concept of texture deformation energy parameters. Finally, merges previously mentioned two parts to form a new expression feature for facial expression recognition. The proposed method is tested in the Cohn-Kanade database and the BHU facial expression database, and the experimental results show the recognition rates of the proposed method comparing with the existing ones increased by 4.5% and 3.9%.
2013, 35(10): 2411-2417.
doi: 10.3724/SP.J.1146.2012.01317
Abstract:
A new blood vessels automatic detection method in fundus image combing adaptive Pulse Coupled Neural Network (PCNN) and maximal categories variance criterion is proposed. In preprocessing, Contrast Limited Adaptive Histogram Equalization (CLAHE) and two-dimensional Gaussian matched filtering are adopted to improve the contrast between blood vessels and background. Then based on simplified PCNN model and maximal categories variance criterion, the preprocessed fundus image is segmented. In image processing, the linking strength of each PCNN neuron is usually a constant. In order to overcome the limitation, pixels Energy Of Laplace (EOL) is chosen as the linking strength of corresponding PCNN neuron, thus PCNN can adjust its linking strengths according to pixel features adaptively. Finally, the final blood vessels detection result is obtained via postprocessing including area filtering and breakpoint connection. The experiments implemented on the Hoover fundus image database show that the method has relatively higher robustness, effectiveness and reliability.
A new blood vessels automatic detection method in fundus image combing adaptive Pulse Coupled Neural Network (PCNN) and maximal categories variance criterion is proposed. In preprocessing, Contrast Limited Adaptive Histogram Equalization (CLAHE) and two-dimensional Gaussian matched filtering are adopted to improve the contrast between blood vessels and background. Then based on simplified PCNN model and maximal categories variance criterion, the preprocessed fundus image is segmented. In image processing, the linking strength of each PCNN neuron is usually a constant. In order to overcome the limitation, pixels Energy Of Laplace (EOL) is chosen as the linking strength of corresponding PCNN neuron, thus PCNN can adjust its linking strengths according to pixel features adaptively. Finally, the final blood vessels detection result is obtained via postprocessing including area filtering and breakpoint connection. The experiments implemented on the Hoover fundus image database show that the method has relatively higher robustness, effectiveness and reliability.
2013, 35(10): 2418-2424.
doi: 10.3724/SP.J.1146.2012.01650
Abstract:
A new low dose CT reconstruction model is proposed under the condition of low signal-to-noise ratio measured data, which are caused by reducing the X-ray source tube current in order to avoid the excessive radiation dose. In the objective function of the model, the logarithm likelihood function under Poisson noise is used as the fidelity functional, and sparse prior of image transform domain coefficients is used as the regularization functional. The fidelity functional is more effective than the additive Gaussian noise model, while the regularization the functional can overcome the ill posed problem of image reconstruction expecially in the low-dose situation. By using the linearized Bregman iteration, the sum minimization scheme is split into one step of quadratic programming with variable coefficient and the other step of the denoising issue. It can accelerate the convergence speed through the variable coefficient calculation in the quadratic programming to approximate the original fidelity term. Experimental results show that this proposed approach can be successfully applied to low-dose fan-beam CT reconstruction and it outperforms some existing algorithms including filter back projection algorithm, maximum likelihood algorithm and classical weighted l2 norm reconstruction algorithm.
A new low dose CT reconstruction model is proposed under the condition of low signal-to-noise ratio measured data, which are caused by reducing the X-ray source tube current in order to avoid the excessive radiation dose. In the objective function of the model, the logarithm likelihood function under Poisson noise is used as the fidelity functional, and sparse prior of image transform domain coefficients is used as the regularization functional. The fidelity functional is more effective than the additive Gaussian noise model, while the regularization the functional can overcome the ill posed problem of image reconstruction expecially in the low-dose situation. By using the linearized Bregman iteration, the sum minimization scheme is split into one step of quadratic programming with variable coefficient and the other step of the denoising issue. It can accelerate the convergence speed through the variable coefficient calculation in the quadratic programming to approximate the original fidelity term. Experimental results show that this proposed approach can be successfully applied to low-dose fan-beam CT reconstruction and it outperforms some existing algorithms including filter back projection algorithm, maximum likelihood algorithm and classical weighted l2 norm reconstruction algorithm.
2013, 35(10): 2425-2431.
doi: 10.3724/SP.J.1146.2012.01171
Abstract:
This paper presents a pattern recognition method for multiclass cancer molecular classification using evolutionary hypernetworks. A multiclass classification issue is decomposed into a set of binary classification issues by One-Versus-All (OVA) approach. The signal-to-noise ratio method is employed for informative genes selection from the DNA microarray. A series of binary classifiers are evolved and used to build a final ensemble classifier for multiclass classification through an evolutionary learning procedure of the hypernetwork. The test sample is classified by using the ensemble classifier. Experimental results show that the Leave One Out Cross Validation (LOOCV) accuracy of the acute leukemia dataset, the small, round blue cell tumor dataset, and the GCM dataset is 98.61%, 100% and 85.35%, respectively. The evolutionary hypernetworks is fit to find cancer-related genes and has a good readability of the learned results.
This paper presents a pattern recognition method for multiclass cancer molecular classification using evolutionary hypernetworks. A multiclass classification issue is decomposed into a set of binary classification issues by One-Versus-All (OVA) approach. The signal-to-noise ratio method is employed for informative genes selection from the DNA microarray. A series of binary classifiers are evolved and used to build a final ensemble classifier for multiclass classification through an evolutionary learning procedure of the hypernetwork. The test sample is classified by using the ensemble classifier. Experimental results show that the Leave One Out Cross Validation (LOOCV) accuracy of the acute leukemia dataset, the small, round blue cell tumor dataset, and the GCM dataset is 98.61%, 100% and 85.35%, respectively. The evolutionary hypernetworks is fit to find cancer-related genes and has a good readability of the learned results.
2013, 35(10): 2432-2437.
doi: 10.3724/SP.J.1146.2012.01311
Abstract:
A new method called generalized Multi-Window S-Transform (MW-ST) is proposed for blind Instantaneous Frequency (IF) analysis of complex emitter signals. The high concentration is achieved by employing the modified Hermite function. In this distribution, the orthogonal Hermite function is improved by utilizing a frequency parameter. Therefore, the multi-window method can be introducing to S-transform. The generalized MW-ST maintains the multi-resolution characteristics of the ST and improves the frequency resolution in high frequency region. Comparing with the origin S transform and the multi-window method, the proposed method can effectively improve the accuracy of the IF estimation and the time-frequency concentration. So it has the good prospects for engineering applications.
A new method called generalized Multi-Window S-Transform (MW-ST) is proposed for blind Instantaneous Frequency (IF) analysis of complex emitter signals. The high concentration is achieved by employing the modified Hermite function. In this distribution, the orthogonal Hermite function is improved by utilizing a frequency parameter. Therefore, the multi-window method can be introducing to S-transform. The generalized MW-ST maintains the multi-resolution characteristics of the ST and improves the frequency resolution in high frequency region. Comparing with the origin S transform and the multi-window method, the proposed method can effectively improve the accuracy of the IF estimation and the time-frequency concentration. So it has the good prospects for engineering applications.
2013, 35(10): 2438-2444.
doi: 10.3724/SP.J.1146.2013.00072
Abstract:
The dynamic task planning issue of early warning system of Low Earth Orbit (LEO) is described by the system resource and missile tracking task aspects, and the Dynamic Constraint Satisfaction Problem (DCSP) model of system dynamic task planning is built which includes two level indexthe integrative optimizing indexes of tracking precision, task accomplishment, sensor switching and resource slack, the adjusting range of original task planning scheme. To solve the dynamic planning model, a variable neighborhood heuristic algorithm on the basis of original planning and schedule is put forward, and the directly insert, redistribute, replace and delete neighborhood structure and their operators are designed in the heuristic algorithm. The simulation results show the leveled dynamic planning model is rational and the heuristic algorithm can solve the dynamic task planning issue of early warning system of LEO effectively.
The dynamic task planning issue of early warning system of Low Earth Orbit (LEO) is described by the system resource and missile tracking task aspects, and the Dynamic Constraint Satisfaction Problem (DCSP) model of system dynamic task planning is built which includes two level indexthe integrative optimizing indexes of tracking precision, task accomplishment, sensor switching and resource slack, the adjusting range of original task planning scheme. To solve the dynamic planning model, a variable neighborhood heuristic algorithm on the basis of original planning and schedule is put forward, and the directly insert, redistribute, replace and delete neighborhood structure and their operators are designed in the heuristic algorithm. The simulation results show the leveled dynamic planning model is rational and the heuristic algorithm can solve the dynamic task planning issue of early warning system of LEO effectively.
2013, 35(10): 2445-2452.
doi: 10.3724/SP.J.1146.2012.01685
Abstract:
Due to the problem of azimuth spectral folding and large migration and strong range dependence on the second range compression for high-squinted spotlight. This paper proposes a method combining the advantages of SPECtral ANalysis (SPECAN) with Extended Non-linear Chirp Scaling (ENCS). First, after Range Cell Migration Correction (RCMC), azimuth spectral folding problem is removed by SPECAN. The variation of quadratic range cell migration correction is removed by chirp scaling. The chirp rates of scatterers in the same range gate are dependent on the azimuth position because of RCMC. So, the extended nonlinear chirp scaling is used to solve the problem of chirp rate dependent on azimuth position. Then focused scope of azimuth is extended. Simulation results verify that this method can get good quality image for high resolution in high squinted spotlight mode.
Due to the problem of azimuth spectral folding and large migration and strong range dependence on the second range compression for high-squinted spotlight. This paper proposes a method combining the advantages of SPECtral ANalysis (SPECAN) with Extended Non-linear Chirp Scaling (ENCS). First, after Range Cell Migration Correction (RCMC), azimuth spectral folding problem is removed by SPECAN. The variation of quadratic range cell migration correction is removed by chirp scaling. The chirp rates of scatterers in the same range gate are dependent on the azimuth position because of RCMC. So, the extended nonlinear chirp scaling is used to solve the problem of chirp rate dependent on azimuth position. Then focused scope of azimuth is extended. Simulation results verify that this method can get good quality image for high resolution in high squinted spotlight mode.
2013, 35(10): 2453-2459.
doi: 10.3724/SP.J.1146.2012.01004
Abstract:
The combination of Frequency Modulation Continuous Wave (FMCW) and Synthetic Aperture Radar (SAR) leads to the rapid development of a compact, low power consumption, cheap and high resolution imaging sensor. A corrected signal model of FMCW SAR is built in this paper which forms the basis for the following study. The generalized expression of FMCW SAR algorithms in frequency domain is derived. For the sake of reduce the computation, a simplified imaging flow with motion compensation operation is proposed and the errors introduced by the simplification is analyzed in detail. Finally, the simulated FMCW SAR data verify the feasibility of proposed method.
The combination of Frequency Modulation Continuous Wave (FMCW) and Synthetic Aperture Radar (SAR) leads to the rapid development of a compact, low power consumption, cheap and high resolution imaging sensor. A corrected signal model of FMCW SAR is built in this paper which forms the basis for the following study. The generalized expression of FMCW SAR algorithms in frequency domain is derived. For the sake of reduce the computation, a simplified imaging flow with motion compensation operation is proposed and the errors introduced by the simplification is analyzed in detail. Finally, the simulated FMCW SAR data verify the feasibility of proposed method.
2013, 35(10): 2460-2466.
doi: 10.3724/SP.J.1146.2012.01689
Abstract:
A novel helicopter-borne Frequency Modulated Continue Wave (FMCW) ROtating Synthetic Aperture Radar (ROSAR) imaging method is proposed. Firstly, by performing the equivalent phase center principle, the separated transmitting and receiving antenna system is equalized to the case of system configuration with antenna both for transmitting and receiving signals. Based on this, accurate two-dimensional spectrum is obtained and the Doppler frequency shift effect induced by the platforms continuous motion while radar transmits and receives signals is analyzed in detail and compensated. Then, efficient inverse Chirp-Z transform is applied to correct the range-variant Range Cell Migration (RCM). With only FFT and complex multiplication and no interpolation, the proposed method can be efficiently implemented. Finally, correctness of the analysis and effectiveness of the proposed algorithm are demonstrated through simulation results.
A novel helicopter-borne Frequency Modulated Continue Wave (FMCW) ROtating Synthetic Aperture Radar (ROSAR) imaging method is proposed. Firstly, by performing the equivalent phase center principle, the separated transmitting and receiving antenna system is equalized to the case of system configuration with antenna both for transmitting and receiving signals. Based on this, accurate two-dimensional spectrum is obtained and the Doppler frequency shift effect induced by the platforms continuous motion while radar transmits and receives signals is analyzed in detail and compensated. Then, efficient inverse Chirp-Z transform is applied to correct the range-variant Range Cell Migration (RCM). With only FFT and complex multiplication and no interpolation, the proposed method can be efficiently implemented. Finally, correctness of the analysis and effectiveness of the proposed algorithm are demonstrated through simulation results.
2013, 35(10): 2467-2474.
doi: 10.3724/SP.J.1146.2012.01534
Abstract:
Carrying out 3-D imaging with multi-aspect SAR data is impressive to radar Automatic Target Recognition (ATR). This paper presents a multi-aspect SAR 3-D imaging technique based on compressive sensing, provides that the target scattering field is sparse. Firstly, it is validated that by multi-aspect SAR measurements the mutual incoherence of measurement matrix is improved. Secondly, the measurement matrix is constructed by carefully selecting the sampling interval in the space domain based on the analysis of mutual incoherence. Finally, the object sparse vector is reconstructed with Stagewise Orthogonal Matching Pursuit (StOMP) algorithm. The proposed method not only improves the resolution of elevation dimension, but also conquers the acute lobe-side resulted from incontinuous sampling. Numerical experiments are given to illustrate the effectiveness and robustness of the proposed method.
Carrying out 3-D imaging with multi-aspect SAR data is impressive to radar Automatic Target Recognition (ATR). This paper presents a multi-aspect SAR 3-D imaging technique based on compressive sensing, provides that the target scattering field is sparse. Firstly, it is validated that by multi-aspect SAR measurements the mutual incoherence of measurement matrix is improved. Secondly, the measurement matrix is constructed by carefully selecting the sampling interval in the space domain based on the analysis of mutual incoherence. Finally, the object sparse vector is reconstructed with Stagewise Orthogonal Matching Pursuit (StOMP) algorithm. The proposed method not only improves the resolution of elevation dimension, but also conquers the acute lobe-side resulted from incontinuous sampling. Numerical experiments are given to illustrate the effectiveness and robustness of the proposed method.
2013, 35(10): 2475-2480.
doi: 10.3724/SP.J.1146.2013.00140
Abstract:
Based on the geometry invariance of rigid target during its 3D motion, the targets unknown 3D shape is reconstructed from 2D images in this paper. Kalman filter and nearest neighbor method data association are used for scattering centers association of different ISAR images. The factorization method is used to reconstruct the target geometry. Furthermore, an error judge standard is proposed to verify the effectiveness of the reconstruction. The reconstruction is realized by the simulation.
Based on the geometry invariance of rigid target during its 3D motion, the targets unknown 3D shape is reconstructed from 2D images in this paper. Kalman filter and nearest neighbor method data association are used for scattering centers association of different ISAR images. The factorization method is used to reconstruct the target geometry. Furthermore, an error judge standard is proposed to verify the effectiveness of the reconstruction. The reconstruction is realized by the simulation.
2013, 35(10): 2481-2486.
doi: 10.3724/SP.J.1146.2012.01309
Abstract:
A novel range-spread target detection algorithm in Gaussian clutter is addressed for Stepped Chirp Modulated Radar (SCMR). For SCMR, the High Resolution Range Profile (HRRP) of a target is obtained by target extraction from overlapping HRRPs which is caused by oversampling. During the target extraction (sometimes called de-correlation), some strong scatters of the target echo are discarded, as the result, the Signal-to-Clutter Ratio (SCR) might be reduced and the target detection capability is degraded. To solve this problem, a novel target detection algorithm without target extraction is addressed. The new algorithm based on the power spectrum of echoes uses not only the amplitude information, but also the phase information of the overlapping HRRPs of a target to improve the SCR, therefore, has significant performance. The test statistic and the false alarm probability of the detector are derived. Finally, the detection performance is assessed by Monte-Carlo simulation, and the results indicate that the proposed algorithm is effective. In addition, the proposed algorithm is robust and easy to implement in practical application.
A novel range-spread target detection algorithm in Gaussian clutter is addressed for Stepped Chirp Modulated Radar (SCMR). For SCMR, the High Resolution Range Profile (HRRP) of a target is obtained by target extraction from overlapping HRRPs which is caused by oversampling. During the target extraction (sometimes called de-correlation), some strong scatters of the target echo are discarded, as the result, the Signal-to-Clutter Ratio (SCR) might be reduced and the target detection capability is degraded. To solve this problem, a novel target detection algorithm without target extraction is addressed. The new algorithm based on the power spectrum of echoes uses not only the amplitude information, but also the phase information of the overlapping HRRPs of a target to improve the SCR, therefore, has significant performance. The test statistic and the false alarm probability of the detector are derived. Finally, the detection performance is assessed by Monte-Carlo simulation, and the results indicate that the proposed algorithm is effective. In addition, the proposed algorithm is robust and easy to implement in practical application.
2013, 35(10): 2487-2492.
doi: 10.3724/SP.J.1146.2012.01149
Abstract:
With prior information of clutter covariance matrix of cells under test, knowledge aided detection methods for distributed targets without secondary data are researched based on Bayesian approach. First, for the heterogeneous clutter environment that the clutter covariance matrix of each cell is not the same with probability one, the Generalized Likelihood Ratio Test (GLRT) and Maximum-A-Posterior (MAP) GLRT are proposed. Then, for the homogeneous clutter environment that the clutter covariance matrix of each cell is the same, the one step GLRT and two-step GLRT are proposed. Furthermore, the detection performance under the prior model mismatched condition is analyzed using computer simulation. The results show that, when the parameter u of prior information model is small, the detection performance of detectors is related to matching degree of prior information. And when the parameter u trends to infinity, all the detectors proposed in this paper have similar performance.
With prior information of clutter covariance matrix of cells under test, knowledge aided detection methods for distributed targets without secondary data are researched based on Bayesian approach. First, for the heterogeneous clutter environment that the clutter covariance matrix of each cell is not the same with probability one, the Generalized Likelihood Ratio Test (GLRT) and Maximum-A-Posterior (MAP) GLRT are proposed. Then, for the homogeneous clutter environment that the clutter covariance matrix of each cell is the same, the one step GLRT and two-step GLRT are proposed. Furthermore, the detection performance under the prior model mismatched condition is analyzed using computer simulation. The results show that, when the parameter u of prior information model is small, the detection performance of detectors is related to matching degree of prior information. And when the parameter u trends to infinity, all the detectors proposed in this paper have similar performance.
2013, 35(10): 2493-2497.
doi: 10.3724/SP.J.1146.2013.00060
Abstract:
Sorting of radar emitters is a vital technology of the radar countermeasure system. To solve the problem that the traditional methods which are based on the Pulse Repetition Interval (PRI) can not sort the signal of PRI complex modulated effectively, this paper introduces a new method which use two-dimensional feature vector that consists of Time Of Arrived (TOA) with frame period to sort radar signals of PRI complex modulated. Even in noisy environment it can get appropriate accuracy of sorting. Extensive Monte Carlo simulation results indicate the effectiveness of the method.
Sorting of radar emitters is a vital technology of the radar countermeasure system. To solve the problem that the traditional methods which are based on the Pulse Repetition Interval (PRI) can not sort the signal of PRI complex modulated effectively, this paper introduces a new method which use two-dimensional feature vector that consists of Time Of Arrived (TOA) with frame period to sort radar signals of PRI complex modulated. Even in noisy environment it can get appropriate accuracy of sorting. Extensive Monte Carlo simulation results indicate the effectiveness of the method.
2013, 35(10): 2498-2504.
doi: 10.3724/SP.J.1146.2012.01614
Abstract:
This paper explores the theory of Compressive Sensing (CS) in radar and evaluates the perturbing effect on measurement noise, channel inference and radar system accuracy error. The performance of traditional Compressive Sensing Radar (CSR) are sensitivity to the above perturbations, which causing the mismatch between non-adaptive random measurement and sensing matrix. To solve the problem, a robust algorithm via Bayesian Compressive Sensing (BCS) with application to noise MIMO radar is proposed. First, a noise MIMO radar sparse sensing model is established and the jointly probability density function based on sparse Bayesian model is derived. Then the BCS algorithm and Least-Absolute Shrinkage and Selection Operator (LASSO) algorithm are employed to optimize the jointly probability density function. Comparing with traditional CSR algorithms, this method estimates effectively the parameters of target when existing mismatch in CSR model, reduces the target information estimation error, and enhances the accuracy and robustness of CSR target information extraction. The validity of the proposed method is illustrated by numerical example.
This paper explores the theory of Compressive Sensing (CS) in radar and evaluates the perturbing effect on measurement noise, channel inference and radar system accuracy error. The performance of traditional Compressive Sensing Radar (CSR) are sensitivity to the above perturbations, which causing the mismatch between non-adaptive random measurement and sensing matrix. To solve the problem, a robust algorithm via Bayesian Compressive Sensing (BCS) with application to noise MIMO radar is proposed. First, a noise MIMO radar sparse sensing model is established and the jointly probability density function based on sparse Bayesian model is derived. Then the BCS algorithm and Least-Absolute Shrinkage and Selection Operator (LASSO) algorithm are employed to optimize the jointly probability density function. Comparing with traditional CSR algorithms, this method estimates effectively the parameters of target when existing mismatch in CSR model, reduces the target information estimation error, and enhances the accuracy and robustness of CSR target information extraction. The validity of the proposed method is illustrated by numerical example.
2013, 35(10): 2505-2511.
doi: 10.3724/SP.J.1146.2013.00005
Abstract:
In bistatic MIMO radar system, all the clutter ridges are located in the same plane of the three- dimensional space. By using this characteristic, a weighted projection optimization based range-dependent clutter suppression method is presented. The coordinates of the clutter spectra are first rotated through a special angle. Then the projection weights are optimized by maximizing the output Signal-Interference-Noise Ratio (SINR) to transform the 3-Dimensional (3D) range-dependent clutter to 2D range-independent. Furthermore, the resultant nonlinear optimization issue is reformulated as a Semi-Definite Programming (SDP) one, which can be solved very efficiently. Analysis of simulation indicate that the proposed method is easy to implement and can cancel the range-dependent clutter effectively.
In bistatic MIMO radar system, all the clutter ridges are located in the same plane of the three- dimensional space. By using this characteristic, a weighted projection optimization based range-dependent clutter suppression method is presented. The coordinates of the clutter spectra are first rotated through a special angle. Then the projection weights are optimized by maximizing the output Signal-Interference-Noise Ratio (SINR) to transform the 3-Dimensional (3D) range-dependent clutter to 2D range-independent. Furthermore, the resultant nonlinear optimization issue is reformulated as a Semi-Definite Programming (SDP) one, which can be solved very efficiently. Analysis of simulation indicate that the proposed method is easy to implement and can cancel the range-dependent clutter effectively.
2013, 35(10): 2512-2517.
doi: 10.3724/SP.J.1146.2013.00121
Abstract:
Among the basic steps of low-altitude wind shear detection for airborne weather radar, the wind speed estimate accuracy is the most important affecting factors. In this paper, a novel method of wind speed estimation based on compressive sensing is proposed to solve the problem of performance degradation in low signal-to-noise ratio and few pulses. According to the sparsity of radar echoes, the Doppler vector is used to design a redundant dictionary for the sparse representation of the signal. The signal compression processing is achieved by using the observation matrix, and then the reconstruction algorithm is used to recover the sparse signal and acquire the accurate estimate of the wind speed. Experimental results show that the proposed method can achieve the accurate wind speed estimation and improve the spectral resolution in low signal-to-noise ratio and few pulses, which means it can identify the wind shear and clutter spectrum even in the adjacent areas.
Among the basic steps of low-altitude wind shear detection for airborne weather radar, the wind speed estimate accuracy is the most important affecting factors. In this paper, a novel method of wind speed estimation based on compressive sensing is proposed to solve the problem of performance degradation in low signal-to-noise ratio and few pulses. According to the sparsity of radar echoes, the Doppler vector is used to design a redundant dictionary for the sparse representation of the signal. The signal compression processing is achieved by using the observation matrix, and then the reconstruction algorithm is used to recover the sparse signal and acquire the accurate estimate of the wind speed. Experimental results show that the proposed method can achieve the accurate wind speed estimation and improve the spectral resolution in low signal-to-noise ratio and few pulses, which means it can identify the wind shear and clutter spectrum even in the adjacent areas.
2013, 35(10): 2518-2523.
doi: 10.3724/SP.J.1146.2013.00033
Abstract:
A spatial semantic model based method is proposed to solve the issue of automatically detecting geo-objects in high resolution remote sensing images. This method obtains firstly image segments through over-segmentation algorithm, and calculates the multiple features by using topic models, in order to improve the description accuracy of segments attribution. Then, this method investigates and models the spatial relationship between geo-objects in whole images, and a semantic parsing tree of the scene category is extracted, which could be used to detect and locate the geo-objects. The experimental results on the dataset demonstrate the robustness and accuracy of this method.
A spatial semantic model based method is proposed to solve the issue of automatically detecting geo-objects in high resolution remote sensing images. This method obtains firstly image segments through over-segmentation algorithm, and calculates the multiple features by using topic models, in order to improve the description accuracy of segments attribution. Then, this method investigates and models the spatial relationship between geo-objects in whole images, and a semantic parsing tree of the scene category is extracted, which could be used to detect and locate the geo-objects. The experimental results on the dataset demonstrate the robustness and accuracy of this method.
2013, 35(10): 2524-2431.
doi: 10.3724/SP.J.1146.2012.01511
Abstract:
A low noise interface circuit on chip is designed for capacitive micro-gyroscope with demodulation signal phase calibration technique. Fully differential trans-impedance amplifier structure is adopted in the readout circuits to optimize the noise performance, which achieves an equivalent input capacitive noise of 0.63 aF/\sqrt{Hz}. Dual channel quadrature demodulation technique and demodulation signal phase calibration technique are simultaneously applied to the sense channel, so that the interference of the mechanical quadrature signal is removed completely. The chip is implemented in 0.35m CMOS process. Experiments on a capacitive gyroscope show that an equivalent input noise of 0.01/s/\sqrt{Hz} is achieved. Under conditions of 8.5 mV//s scale factor and 5 V supply voltage, the full scale range of the gyroscope reaches 200/swith only 2 nonlinearity.
A low noise interface circuit on chip is designed for capacitive micro-gyroscope with demodulation signal phase calibration technique. Fully differential trans-impedance amplifier structure is adopted in the readout circuits to optimize the noise performance, which achieves an equivalent input capacitive noise of 0.63 aF/\sqrt{Hz}. Dual channel quadrature demodulation technique and demodulation signal phase calibration technique are simultaneously applied to the sense channel, so that the interference of the mechanical quadrature signal is removed completely. The chip is implemented in 0.35m CMOS process. Experiments on a capacitive gyroscope show that an equivalent input noise of 0.01/s/\sqrt{Hz} is achieved. Under conditions of 8.5 mV//s scale factor and 5 V supply voltage, the full scale range of the gyroscope reaches 200/swith only 2 nonlinearity.
2013, 35(10): 2532-2535.
doi: 10.3724/SP.J.1146.2012.01746
Abstract:
Multiplied by constant on modulo 2n operation(y=cx mod 2n), is widely used in the ciphers like Sosemanuk, RC6, MARS, and so on. This operation is recognized as a permutation with considerable diffusion, confusion and fine realization efficiency, where the constant c is odd. The operation can be viewed as a vector Boolean function, which vector Walsh spectrum character is not analyzed in published paper. In this paper, the property of the vector Walsh spectrum distribution of the operation is studied, the structure and counting formulas of input and output linear masks and the constant are given for the first time, where the Walsh spectrum of the operation is to be 1. It is proved that there is not input and output linear masks which Walsh spectrum is to be -1.
Multiplied by constant on modulo 2n operation(y=cx mod 2n), is widely used in the ciphers like Sosemanuk, RC6, MARS, and so on. This operation is recognized as a permutation with considerable diffusion, confusion and fine realization efficiency, where the constant c is odd. The operation can be viewed as a vector Boolean function, which vector Walsh spectrum character is not analyzed in published paper. In this paper, the property of the vector Walsh spectrum distribution of the operation is studied, the structure and counting formulas of input and output linear masks and the constant are given for the first time, where the Walsh spectrum of the operation is to be 1. It is proved that there is not input and output linear masks which Walsh spectrum is to be -1.
2013, 35(10): 2536-2540.
doi: 10.3724/SP.J.1146.2012.01574
Abstract:
Vaudenay (1999) proved that the permutation in Lai-Massey scheme should be an orthomorphism or almost orthomorphism. This paper mainly focuses on the principle of the function in Lai-Massey scheme, which is described by its resistance to differential and linear attack. It shows that no matter how the group G is defined, ifis an affine function on G, then it should be defined as an orthomorphism, or else there exists a differentially characteristic with probability 1 and a linearly approximation with correlation coefficient 1, therefore it has potential security risk. Moreover, by the characteristic spectrum in finite group, a new linear relationship between the input and output of Lai-Massey scheme is introduced, which is used to describe the linear relationship lying between the input and the output of Lai-Massey scheme.
Vaudenay (1999) proved that the permutation in Lai-Massey scheme should be an orthomorphism or almost orthomorphism. This paper mainly focuses on the principle of the function in Lai-Massey scheme, which is described by its resistance to differential and linear attack. It shows that no matter how the group G is defined, ifis an affine function on G, then it should be defined as an orthomorphism, or else there exists a differentially characteristic with probability 1 and a linearly approximation with correlation coefficient 1, therefore it has potential security risk. Moreover, by the characteristic spectrum in finite group, a new linear relationship between the input and output of Lai-Massey scheme is introduced, which is used to describe the linear relationship lying between the input and the output of Lai-Massey scheme.
