Email alert
2012 Vol. 34, No. 4
Display Method:
2012, 34(4): 763-769.
doi: 10.3724/SP.J.1146.2011.01368
Abstract:
Addressing scheme for Internet of Things (IoTs) is proposed based on IPv6 over Low-power Wireless Personal Area Networks (6LoWPAN), for the real-time communication?between Internet and underlying heterogeneous networks based on IEEE 802.15.4. Addressing scheme includes IPv6 address autoconfiguration and head compression. Hierarchical address autoconfiguration firstly allow nodes use link local address, which is derived by 16-bit short address, for transmitting data packets within WPAN, meanwhile, this link local address needs to ensure the uniqueness of 16-bit short address by performing duplicate address detection based on clustering. Secondly, all the sink nodes in underlying networks form global address by obtaining global address prefix, which combine with interface identifier, and achieve data exchange between Internet and the underlying networks. Simultaneity, IoTs IPv6 Header Compression (IIPHC) is proposed by embedding G/L bit in head compression encoding. For the link local address, simple IIPHC1 program is used. For the global address, IIPHC2 that is relatively complex but effective program is used. The simulation result indicates that addressing scheme have an improvement in networks overhead, latency, throughput and energy consumption.
Addressing scheme for Internet of Things (IoTs) is proposed based on IPv6 over Low-power Wireless Personal Area Networks (6LoWPAN), for the real-time communication?between Internet and underlying heterogeneous networks based on IEEE 802.15.4. Addressing scheme includes IPv6 address autoconfiguration and head compression. Hierarchical address autoconfiguration firstly allow nodes use link local address, which is derived by 16-bit short address, for transmitting data packets within WPAN, meanwhile, this link local address needs to ensure the uniqueness of 16-bit short address by performing duplicate address detection based on clustering. Secondly, all the sink nodes in underlying networks form global address by obtaining global address prefix, which combine with interface identifier, and achieve data exchange between Internet and the underlying networks. Simultaneity, IoTs IPv6 Header Compression (IIPHC) is proposed by embedding G/L bit in head compression encoding. For the link local address, simple IIPHC1 program is used. For the global address, IIPHC2 that is relatively complex but effective program is used. The simulation result indicates that addressing scheme have an improvement in networks overhead, latency, throughput and energy consumption.
2012, 34(4): 770-775.
doi: 10.3724/SP.J.1146.2011.01002
Abstract:
This paper studies on the issue of physical layer security of MISO Cognitive Radio Networks (CRN) from the Quality of Service (QoS) perspective. By adding a suitable amount of artificial noise into the transmitting signal of the Secondary User Transmitter (SU-Tx), the security performance could be enhanced apparently. Through some appropriate transformations, the optimization problem is converted into a semi-definite programming, which can be easily solved. Furthermore, a robust optimal transmitter is designed by using the worst-case optimization approach on the premise of knowing the region of uncertainties of Channel State Information (CSI). Therefore, all the constraints can still be satisfied even with the maximal CSI errors. Simulation results verify the effectiveness of the proposed approach.
This paper studies on the issue of physical layer security of MISO Cognitive Radio Networks (CRN) from the Quality of Service (QoS) perspective. By adding a suitable amount of artificial noise into the transmitting signal of the Secondary User Transmitter (SU-Tx), the security performance could be enhanced apparently. Through some appropriate transformations, the optimization problem is converted into a semi-definite programming, which can be easily solved. Furthermore, a robust optimal transmitter is designed by using the worst-case optimization approach on the premise of knowing the region of uncertainties of Channel State Information (CSI). Therefore, all the constraints can still be satisfied even with the maximal CSI errors. Simulation results verify the effectiveness of the proposed approach.
2012, 34(4): 776-781.
doi: 10.3724/SP.J.1146.2011.00653
Abstract:
A dynamic offset time assignment algorithm based on the measurement of cross traffic on control plane is proposed. This algorithm uses probe burst,s drop ratio as observing variable to determine the core node,s background traffic and assigns offset time dynamically based on the measured result. Analysis and simulation show that this algorithm can get moderate offset time under target Insufficient Offset Time (IOT) drop ratio,s restriction and obtain tradeoff between end to end delay and IOT ratio.
A dynamic offset time assignment algorithm based on the measurement of cross traffic on control plane is proposed. This algorithm uses probe burst,s drop ratio as observing variable to determine the core node,s background traffic and assigns offset time dynamically based on the measured result. Analysis and simulation show that this algorithm can get moderate offset time under target Insufficient Offset Time (IOT) drop ratio,s restriction and obtain tradeoff between end to end delay and IOT ratio.
2012, 34(4): 782-786.
doi: 10.3724/SP.J.1146.2011.00301
Abstract:
User clustering is one of the most important problems because of the difference of the frequency spectrum utilization situation from cognitive users. This paper gives the analysis result of correlation between cognitive radio users and also proposes the clustering algorithm based on this analysis. Considering the real situation, the effects of data quantization are derived and the derivation shows the performance loss could be compensated partly through the increase the number of spectrum bands. Finally, the simulation shows the proposal could perform well whether the data quantization is adopted or not. From the aspect of reliability, accuracy and adaptability, the proposed algorithm, which gives a comprehensive consideration of the spectrum environment and other factors, is more practical than the traditional clustering algorithm based on the geographic location.
User clustering is one of the most important problems because of the difference of the frequency spectrum utilization situation from cognitive users. This paper gives the analysis result of correlation between cognitive radio users and also proposes the clustering algorithm based on this analysis. Considering the real situation, the effects of data quantization are derived and the derivation shows the performance loss could be compensated partly through the increase the number of spectrum bands. Finally, the simulation shows the proposal could perform well whether the data quantization is adopted or not. From the aspect of reliability, accuracy and adaptability, the proposed algorithm, which gives a comprehensive consideration of the spectrum environment and other factors, is more practical than the traditional clustering algorithm based on the geographic location.
2012, 34(4): 787-794.
doi: 10.3724/SP.J.1146.2011.00554
Abstract:
Most existing works on resource allocation in Orthogonal Frequency Division Multiple Access (OFDMA) relaying networks is focused on single cell systems, which ignoring the significant effect of co-channel interference caused by adjacent cells. However, in practice, the higher frequency reuse factor and small cell size requirement lead to severe inter-cell interference problem. In this paper, the resource allocation in multi-cell downlink OFDMA decode-and-forward relaying networks is considered. The problem has a mixed discrete programming structure and is known to be NP-hard even for single cell scenarios. A distributed suboptimal resource allocation scheme is then developed due to the inherent complexity of implementing the optimal solution. The proposed scheme is performed in two steps: firstly the subcarriers are allocated to subscribers to provide QoS continuity requirements as well as significantly reducing the network signaling. Then the power control problem is approximately transformed and decomposed into smaller convex optimization subproblems whose solutions are jointly and iteratively coordinated by the use of dual variables based on the ellipsoid method. Simulation results show that the proposed scheme outperforms the reference schemes, in terms of system capacity and cell edge throughput.
Most existing works on resource allocation in Orthogonal Frequency Division Multiple Access (OFDMA) relaying networks is focused on single cell systems, which ignoring the significant effect of co-channel interference caused by adjacent cells. However, in practice, the higher frequency reuse factor and small cell size requirement lead to severe inter-cell interference problem. In this paper, the resource allocation in multi-cell downlink OFDMA decode-and-forward relaying networks is considered. The problem has a mixed discrete programming structure and is known to be NP-hard even for single cell scenarios. A distributed suboptimal resource allocation scheme is then developed due to the inherent complexity of implementing the optimal solution. The proposed scheme is performed in two steps: firstly the subcarriers are allocated to subscribers to provide QoS continuity requirements as well as significantly reducing the network signaling. Then the power control problem is approximately transformed and decomposed into smaller convex optimization subproblems whose solutions are jointly and iteratively coordinated by the use of dual variables based on the ellipsoid method. Simulation results show that the proposed scheme outperforms the reference schemes, in terms of system capacity and cell edge throughput.
2012, 34(4): 795-801.
doi: 10.3724/SP.J.1146.2011.00636
Abstract:
In asynchronous MIMO-OFDM over the selective Rayleigh fading channel, an iterative parallel multi- antenna interference cancellation algorithm based on pre-processing matrix is proposed. Before the signal is transmitted, the symbols are spread over all the subcarriers, which relieve the influence of deep fading in part of the subcarriers. In the receiver, the wrong decision value of last iteration is spread by the pre-processing matrix, which decrease the error propagation in iteration processing. The performance improvement is verified by simulation, when there are 4 transmitters 2 receivers and the Bit Error Rate (BER) is 10-5, SNR is improved about 4.5 dB compared with the traditional iterative parallel interference cancellation algorithm.
In asynchronous MIMO-OFDM over the selective Rayleigh fading channel, an iterative parallel multi- antenna interference cancellation algorithm based on pre-processing matrix is proposed. Before the signal is transmitted, the symbols are spread over all the subcarriers, which relieve the influence of deep fading in part of the subcarriers. In the receiver, the wrong decision value of last iteration is spread by the pre-processing matrix, which decrease the error propagation in iteration processing. The performance improvement is verified by simulation, when there are 4 transmitters 2 receivers and the Bit Error Rate (BER) is 10-5, SNR is improved about 4.5 dB compared with the traditional iterative parallel interference cancellation algorithm.
2012, 34(4): 802-806.
doi: 10.3724/SP.J.1146.2011.00562
Abstract:
This paper studies single-cycle polynomials over the integer residue ring Z/(pn) with prime p5 and integer n2, and presents several classes of such single-cycle polynomials. As the research of single-cycle polynomials over Z/(pn) can be reduced to the case over Z/(p2), an exact characterization of single-cycle polynomials over Z/(5) is given in terms of their coefficients, and then a complete characterization of single-cycle polynomials of degree 6 over Z/(52) is given based on it. In addition, a partial construction of single-cycle polynomials of degree (p1) over Z/(p2) is also proposed.
This paper studies single-cycle polynomials over the integer residue ring Z/(pn) with prime p5 and integer n2, and presents several classes of such single-cycle polynomials. As the research of single-cycle polynomials over Z/(pn) can be reduced to the case over Z/(p2), an exact characterization of single-cycle polynomials over Z/(5) is given in terms of their coefficients, and then a complete characterization of single-cycle polynomials of degree 6 over Z/(52) is given based on it. In addition, a partial construction of single-cycle polynomials of degree (p1) over Z/(p2) is also proposed.
2012, 34(4): 807-811.
doi: 10.3724/SP.J.1146.2011.00863
Abstract:
Stream cipher HC-256' is an improved algorithm of HC-256 which is proposed as a candidate to the eSTREAM project. Until now, there has not any cryptanalysis on HC-256'. In this paper, a linear distinguishing attack on HC-256' is presented. This method uses different nonlinear functions instead of state update functions to exploit the weaknesses in the even positions output bits of the keystream generation sequence. By linear approximation to the internal state bits, a distinguisher is built. The result shows that there needs about2281 bit keystream with advantage 0.9545 to distinguish HC-256' form random sequence. Thereby, this is a beneficial attempt to solve a problem which is given by Sekar et al in IWSEC 2009.
Stream cipher HC-256' is an improved algorithm of HC-256 which is proposed as a candidate to the eSTREAM project. Until now, there has not any cryptanalysis on HC-256'. In this paper, a linear distinguishing attack on HC-256' is presented. This method uses different nonlinear functions instead of state update functions to exploit the weaknesses in the even positions output bits of the keystream generation sequence. By linear approximation to the internal state bits, a distinguisher is built. The result shows that there needs about2281 bit keystream with advantage 0.9545 to distinguish HC-256' form random sequence. Thereby, this is a beneficial attempt to solve a problem which is given by Sekar et al in IWSEC 2009.
2012, 34(4): 812-817.
doi: 10.3724/SP.J.1146.2011.00787
Abstract:
Considering at the trust problem existing in cloud computing environment, a double incentive based on trust and deception detection model CCIDTM (Cloud Computing Incentive and Detection Trust Model) is proposed. The model proposes a set of cloud computing services property evaluation, introducing a dynamic mechanism of trust decaying over time, establishing a double incentive mechanism about service behavior of service providers as well as users evaluate behavior. An algorithm of conspiracy to deceive group testing which improves the model of dynamic adaptation and evaluation of comprehensive is also proposed. Experiment results show that the model results of the assessment service providers closed to the true service behavior. Thus it can effectively resist the attacks of malicious behavior, showing a good robustness compared with the existing trust model.
Considering at the trust problem existing in cloud computing environment, a double incentive based on trust and deception detection model CCIDTM (Cloud Computing Incentive and Detection Trust Model) is proposed. The model proposes a set of cloud computing services property evaluation, introducing a dynamic mechanism of trust decaying over time, establishing a double incentive mechanism about service behavior of service providers as well as users evaluate behavior. An algorithm of conspiracy to deceive group testing which improves the model of dynamic adaptation and evaluation of comprehensive is also proposed. Experiment results show that the model results of the assessment service providers closed to the true service behavior. Thus it can effectively resist the attacks of malicious behavior, showing a good robustness compared with the existing trust model.
2012, 34(4): 818-824.
doi: 10.3724/SP.J.1146.2011.00823
Abstract:
This paper focuses on the issue of available bandwidth estimation in wireless multihop path and takes the overall QoS provision as the principle in estimation. Through analyzing and quantitative calculating the interference on the multihop path, the analytical model for performance analysis is derived based on queueing networks theory and from which the QoS parameters of each flow along the path are obtained. Based on the model a method for available bandwidth estimation for QoS provision is proposed, which takes the flow,s QoS demands as the constraints and thus overcomes the drawbacks of the existing schemes that only take the maximum achievable throughput as the available bandwidth while the flows other QoS demands may be affected. Both the proposed model and the estimation method are validated through extensive simulations.
This paper focuses on the issue of available bandwidth estimation in wireless multihop path and takes the overall QoS provision as the principle in estimation. Through analyzing and quantitative calculating the interference on the multihop path, the analytical model for performance analysis is derived based on queueing networks theory and from which the QoS parameters of each flow along the path are obtained. Based on the model a method for available bandwidth estimation for QoS provision is proposed, which takes the flow,s QoS demands as the constraints and thus overcomes the drawbacks of the existing schemes that only take the maximum achievable throughput as the available bandwidth while the flows other QoS demands may be affected. Both the proposed model and the estimation method are validated through extensive simulations.
2012, 34(4): 825-831.
doi: 10.3724/SP.J.1146.2011.00446
Abstract:
Based on probabilistic sensing model, a virtual sensor is constructed by information fusion scheme that fuses sensing measurements of two sensors, which can significantly improve coverage region. Based on the theories, a barrier coverage preserving configuration algorithm is proposed. The algorithm provides continuous barrier coverage for the whole region by divide-and-conquer approach, which can reduce the communication overhead. Moreover, the redundancy nodes are scheduled to sleep for reducing energy consumption and prolonging the network lifetime. Analysis and experiment results show that the validity of model and algorithm enhancing coverage range and increasing the max separation distance between two sensors, and the algorithm outperforms the barrier coverage preserving configuration scheme based on information non-fusion sensor, on the number of barrier and network lifetime.
Based on probabilistic sensing model, a virtual sensor is constructed by information fusion scheme that fuses sensing measurements of two sensors, which can significantly improve coverage region. Based on the theories, a barrier coverage preserving configuration algorithm is proposed. The algorithm provides continuous barrier coverage for the whole region by divide-and-conquer approach, which can reduce the communication overhead. Moreover, the redundancy nodes are scheduled to sleep for reducing energy consumption and prolonging the network lifetime. Analysis and experiment results show that the validity of model and algorithm enhancing coverage range and increasing the max separation distance between two sensors, and the algorithm outperforms the barrier coverage preserving configuration scheme based on information non-fusion sensor, on the number of barrier and network lifetime.
2012, 34(4): 832-837.
doi: 10.3724/SP.J.1146.2011.00859
Abstract:
To improve the speed of the estimation of Direction Of Arrival (DOA), the dimension of the noise subspace is descended by the Singular Value decomposing (SVD) on the intersection of noise subspace and its conjugate one. Then a new method for fast 2-D DOA estimation is proposed based on the double orthogonality of the descended noise subspace to the steering vector and its conjugate one. Theoretical analysis and experiment results show that the newly developed method can be used without any restriction by the array structure and is capable of compressing the range of the dimension of traditional MUltiple SIgnal Classification (MUSIC) spectrum for 2 times, therefore, the calculation capacity of DOA estimate can be reduced to 50% while the estimation precision is the same as that of MUSIC.
To improve the speed of the estimation of Direction Of Arrival (DOA), the dimension of the noise subspace is descended by the Singular Value decomposing (SVD) on the intersection of noise subspace and its conjugate one. Then a new method for fast 2-D DOA estimation is proposed based on the double orthogonality of the descended noise subspace to the steering vector and its conjugate one. Theoretical analysis and experiment results show that the newly developed method can be used without any restriction by the array structure and is capable of compressing the range of the dimension of traditional MUltiple SIgnal Classification (MUSIC) spectrum for 2 times, therefore, the calculation capacity of DOA estimate can be reduced to 50% while the estimation precision is the same as that of MUSIC.
2012, 34(4): 838-843.
doi: 10.3724/SP.J.1146.2011.00620
Abstract:
Chaotic Compressive Sensing (ChaCS) is a nonlinear compressive sensing approach using chaos systems. This paper extends the ChaCS to perform the online estimation of sparse time-varying signals. An online estimation structure is proposed and a sparsity-constrained recursive least-squares objective function is formulated. The sparse time-varying signals are estimated through iterative reweighted nonlinear least-square algorithm by minimizing the objective function. The Henon system is taken as examples to expose the estimation performance of frequency sparse time-varying signals. Numerical simulations illustrate the effectiveness of the proposed method.
Chaotic Compressive Sensing (ChaCS) is a nonlinear compressive sensing approach using chaos systems. This paper extends the ChaCS to perform the online estimation of sparse time-varying signals. An online estimation structure is proposed and a sparsity-constrained recursive least-squares objective function is formulated. The sparse time-varying signals are estimated through iterative reweighted nonlinear least-square algorithm by minimizing the objective function. The Henon system is taken as examples to expose the estimation performance of frequency sparse time-varying signals. Numerical simulations illustrate the effectiveness of the proposed method.
2012, 34(4): 844-850.
doi: 10.3724/SP.J.1146.2011.00667
Abstract:
Cellular Automata (CA) and Cellular Automata Transfrom (CAT) are introduced, and a novel watermark algorithm joint JPEG image encoding and cellular automata is proposed. An image watermark is scrambled by Moore CA firstly. And then is embedded into the low-frequency subband, which is obtained after the transform of original image with CAT. Then the watermarked image is encoded according the JPEG image compression standard. The watermark is extracted in the decoding process. The experiments show that the proposed algorithm is robust to the common watermark attacks, such as JPEG compression attack, filter attack, rotation attack, and additive noise attack and so on.
Cellular Automata (CA) and Cellular Automata Transfrom (CAT) are introduced, and a novel watermark algorithm joint JPEG image encoding and cellular automata is proposed. An image watermark is scrambled by Moore CA firstly. And then is embedded into the low-frequency subband, which is obtained after the transform of original image with CAT. Then the watermarked image is encoded according the JPEG image compression standard. The watermark is extracted in the decoding process. The experiments show that the proposed algorithm is robust to the common watermark attacks, such as JPEG compression attack, filter attack, rotation attack, and additive noise attack and so on.
2012, 34(4): 851-857.
doi: 10.3724/SP.J.1146.2011.00851
Abstract:
Based on the similarity measure of evidence, a new method for combining conflicting interval evidence is proposed. Firstly, interval evidence can be transformed into interval-valued Pignistic probability by using the defined extended Pignistic probability function. Using the normalized Euclidean distance of interval-valued fuzzy sets, the similarity between Pignistic probabilities of interval evidence are obtained, and similarity measure matrix can be constructed, from which the credibility degrees (weights) of interval evidence can be got. Secondly, based on the credibility degrees, new interval evidence can be obtained by modified and weightedly averaging the original interval evidence. Using Demspter interval evidence combination rule, the fusion result can be obtained by combining the new interval evidence. The proposed method can effectively eliminate the effect of highly conflicting interval evidence in combination so as to reduce the width of combined interval evidence. Therefore the uncertainty of decision-making can be decreased. Finally, in classical numerical examples, compared with the fused results by directly using Demspter interval evidence combination rule, the combined results by using this proposed method are more rational and reliable.
Based on the similarity measure of evidence, a new method for combining conflicting interval evidence is proposed. Firstly, interval evidence can be transformed into interval-valued Pignistic probability by using the defined extended Pignistic probability function. Using the normalized Euclidean distance of interval-valued fuzzy sets, the similarity between Pignistic probabilities of interval evidence are obtained, and similarity measure matrix can be constructed, from which the credibility degrees (weights) of interval evidence can be got. Secondly, based on the credibility degrees, new interval evidence can be obtained by modified and weightedly averaging the original interval evidence. Using Demspter interval evidence combination rule, the fusion result can be obtained by combining the new interval evidence. The proposed method can effectively eliminate the effect of highly conflicting interval evidence in combination so as to reduce the width of combined interval evidence. Therefore the uncertainty of decision-making can be decreased. Finally, in classical numerical examples, compared with the fused results by directly using Demspter interval evidence combination rule, the combined results by using this proposed method are more rational and reliable.
2012, 34(4): 858-864.
doi: 10.3724/SP.J.1146.2011.00827
Abstract:
In order to improve accuracy of the human detection, this paper proposes the conception of the Window edge of the Gradient of Potential Energy (WGPE) and a fast human detection method based on potential energy. By using sparse-dense gradient potential windows set, the detection time of the multi-scale detection can be shortened. Cascading Support Vector Machine (SVM) training using weighted positive and negative samples, the occlusion sample of the human body is weighted to detect the human body under occlusion. Filter positive in the detection window, the algorithm does not require too much computational overhead increases when the detection window is filtered. In the smooth background image, the proposed method compared to the multi-level Histograms of Oriented Gradients (HOG) detection and HOG-LBP (Local Binary Pattern) methods accuracy at the same rate, spents less testing time. Experiments show that the human detection accuracy and efficiency has increased, the case for the human body in partial occlusion detection, the accuracy rate is improved markedly.
In order to improve accuracy of the human detection, this paper proposes the conception of the Window edge of the Gradient of Potential Energy (WGPE) and a fast human detection method based on potential energy. By using sparse-dense gradient potential windows set, the detection time of the multi-scale detection can be shortened. Cascading Support Vector Machine (SVM) training using weighted positive and negative samples, the occlusion sample of the human body is weighted to detect the human body under occlusion. Filter positive in the detection window, the algorithm does not require too much computational overhead increases when the detection window is filtered. In the smooth background image, the proposed method compared to the multi-level Histograms of Oriented Gradients (HOG) detection and HOG-LBP (Local Binary Pattern) methods accuracy at the same rate, spents less testing time. Experiments show that the human detection accuracy and efficiency has increased, the case for the human body in partial occlusion detection, the accuracy rate is improved markedly.
2012, 34(4): 865-870.
doi: 10.3724/SP.J.1146.2011.00858
Abstract:
A tracking algorithm with adaptive association gate is proposed to the maneuvering target tracking in clutters. The algorithm is based on conventional Interacting Multiple Model Probabilistic Data Association (IMM-PDA) algorithm, and the target is assumed to change its moving mode with the maximal maneuvering level at current or heretofore moments when there is no valid measurement in the association gate. The innovation covariance used to determine the association gate is modified according to the predicted measurement in the maximal maneuvering hypothesis and the actual one. The association gate is enlarged step by step appropriately to gate the target measurement as far as possible. Simulation results show that, the proposed algorithm can improve the tracking loss rates of maneuvering target effectively without decreasing tracking precision or increasing computational complexity.
A tracking algorithm with adaptive association gate is proposed to the maneuvering target tracking in clutters. The algorithm is based on conventional Interacting Multiple Model Probabilistic Data Association (IMM-PDA) algorithm, and the target is assumed to change its moving mode with the maximal maneuvering level at current or heretofore moments when there is no valid measurement in the association gate. The innovation covariance used to determine the association gate is modified according to the predicted measurement in the maximal maneuvering hypothesis and the actual one. The association gate is enlarged step by step appropriately to gate the target measurement as far as possible. Simulation results show that, the proposed algorithm can improve the tracking loss rates of maneuvering target effectively without decreasing tracking precision or increasing computational complexity.
2012, 34(4): 871-877.
doi: 10.3724/SP.J.1146.2011.00796
Abstract:
This paper focus on the detection of floating small targets in high range resolution sea clutter. Floationg targets disarrange the scattering of neighboring sea surface, which results in that the received echoes in the cell targets located satisfy a non-additive model. While, it is hardly to model the paramters correlated to targets in the non-additive model. In order to keep away from the parameter modeling, target detection can be regarded as a binary-classification, where the clutter-only pattern is available for the classifier design and target detection is to judge whether the received echoes belong to the clutter-only pattern. For the classification, a feature united detection algrithm based on the non-additive model is proposed in the paper. First, two extracted features from the received echoes are combined into a normalized vector for target detection. Then, a convex hull training algorithm is utilized to determine a decision region. Finally, the detection rule is whether the decision region surrounds the vector. Experimental results by the raw IPIX radar data show that the proposed algorithm outperforms the compared algorithms. It provides a new detection guidance for the marine radar to detect samll targets.
This paper focus on the detection of floating small targets in high range resolution sea clutter. Floationg targets disarrange the scattering of neighboring sea surface, which results in that the received echoes in the cell targets located satisfy a non-additive model. While, it is hardly to model the paramters correlated to targets in the non-additive model. In order to keep away from the parameter modeling, target detection can be regarded as a binary-classification, where the clutter-only pattern is available for the classifier design and target detection is to judge whether the received echoes belong to the clutter-only pattern. For the classification, a feature united detection algrithm based on the non-additive model is proposed in the paper. First, two extracted features from the received echoes are combined into a normalized vector for target detection. Then, a convex hull training algorithm is utilized to determine a decision region. Finally, the detection rule is whether the decision region surrounds the vector. Experimental results by the raw IPIX radar data show that the proposed algorithm outperforms the compared algorithms. It provides a new detection guidance for the marine radar to detect samll targets.
2012, 34(4): 878-884.
doi: 10.3724/SP.J.1146.2011.00788
Abstract:
This paper presents a novel method for joint parameter estimation of Linear Frequency Modulation (LFM) signals in bistatic Multiple-Input Multiple-Output (MIMO) radar system. A new signal array model of bistatic MIMO radar system is constructed in this paper. Firstly, each LFM signal is extracted according to the characteristic energy concentration in FRactional Fourier Transform (FRFT) domain. Secondly, the Doppler frequency shift scale and time delay are estimated by searching peak point in the FRFT domain. Furthermore, both Direction Of Departures (DODs) and Direction Of Arrivals (DOAs) of the multiple targets are estimated and paired automatically by employing the FRFT-MUSIC algorithm. Simulation results are presented to verify the effectiveness of the proposed method.
This paper presents a novel method for joint parameter estimation of Linear Frequency Modulation (LFM) signals in bistatic Multiple-Input Multiple-Output (MIMO) radar system. A new signal array model of bistatic MIMO radar system is constructed in this paper. Firstly, each LFM signal is extracted according to the characteristic energy concentration in FRactional Fourier Transform (FRFT) domain. Secondly, the Doppler frequency shift scale and time delay are estimated by searching peak point in the FRFT domain. Furthermore, both Direction Of Departures (DODs) and Direction Of Arrivals (DOAs) of the multiple targets are estimated and paired automatically by employing the FRFT-MUSIC algorithm. Simulation results are presented to verify the effectiveness of the proposed method.
2012, 34(4): 885-890.
doi: 10.3724/SP.J.1146.2011.00687
Abstract:
Traditional Dynamic Programming Track-Before-Detect (DP-TBD) algorithms use only observation data of current frame to associate with merit function and accumulate energy at each stage of data association. The ignorance of targets state relevance among successive frames and its own kinematic characters results in false state association at low Signal-to-Noise Ratio (SNR), which reduce detecting and tracking performance profoundly. To solve this issue, a DP-TBD algorithm based on second order Markov target state model is proposed. Taking maximum of the targets state conditional PDF ratio as the optimal criteria, this algorithm makes use of second order Markov model to describe the targets state relevance and defines a state transition probability model according to targets kinematic characters, which relates to targets turning angle. On these bases, a multi-frame data association DP-TBD algorithm is implemented. Compared to traditional DP-TBD algorithm through a simulation experiment, the proposed algorithm turns out to have better detection and tracking performance.
Traditional Dynamic Programming Track-Before-Detect (DP-TBD) algorithms use only observation data of current frame to associate with merit function and accumulate energy at each stage of data association. The ignorance of targets state relevance among successive frames and its own kinematic characters results in false state association at low Signal-to-Noise Ratio (SNR), which reduce detecting and tracking performance profoundly. To solve this issue, a DP-TBD algorithm based on second order Markov target state model is proposed. Taking maximum of the targets state conditional PDF ratio as the optimal criteria, this algorithm makes use of second order Markov model to describe the targets state relevance and defines a state transition probability model according to targets kinematic characters, which relates to targets turning angle. On these bases, a multi-frame data association DP-TBD algorithm is implemented. Compared to traditional DP-TBD algorithm through a simulation experiment, the proposed algorithm turns out to have better detection and tracking performance.
2012, 34(4): 891-897.
doi: 10.3724/SP.J.1146.2011.01020
Abstract:
Speckle makes High Resolution Range Profile (HRRP) fluctuate largely within the frame without scatterer Moving Through Range Cell (MTRC). This paper makes use of the matching score to measure the speckles influence on the spectra features of HRRP, such as the amplitude spectra, power spectra and higher order spectra. Then, the results are analyzed to obtain some conclusions for the selection of spectra features. The Monte Carlo experiments and measured data from an anechoic chamber indicate that the speckle makes spectra features of HRRP fluctuate according to the energy ratio of the speckled scatterers and the speckled segment. Meanwhile, the fluctuation of higher order spectra caused by the speckle also depends on the order. In the common spectra features of HRRP, power spectra are the most insensitive to the speckle.
Speckle makes High Resolution Range Profile (HRRP) fluctuate largely within the frame without scatterer Moving Through Range Cell (MTRC). This paper makes use of the matching score to measure the speckles influence on the spectra features of HRRP, such as the amplitude spectra, power spectra and higher order spectra. Then, the results are analyzed to obtain some conclusions for the selection of spectra features. The Monte Carlo experiments and measured data from an anechoic chamber indicate that the speckle makes spectra features of HRRP fluctuate according to the energy ratio of the speckled scatterers and the speckled segment. Meanwhile, the fluctuation of higher order spectra caused by the speckle also depends on the order. In the common spectra features of HRRP, power spectra are the most insensitive to the speckle.
2012, 34(4): 898-903.
doi: 10.3724/SP.J.1146.2011.00861
Abstract:
To reduce the computational complexity and training samples required of Adaptive Matrix Approach (AMA), a Two-Sided AMA (TS-AMA) for MIMO radar is proposed. The proposed algorithm converts the cost function of AMA into a bi-quadratic one by decomposing the weight matrix of AMA into a Kronecker of two small dimensional weight matrices. The new cost function can be efficiently solved by combining Semi-Definite Programming (SDP) with Bi-Iterative Algorithm (BIA). The proposed algorithm has faster convergence rate, smaller training samples required and lower computational complexity comparable with that of AMA. The numerical examples are provided to demonstrate the effectiveness of the proposed algorithm.
To reduce the computational complexity and training samples required of Adaptive Matrix Approach (AMA), a Two-Sided AMA (TS-AMA) for MIMO radar is proposed. The proposed algorithm converts the cost function of AMA into a bi-quadratic one by decomposing the weight matrix of AMA into a Kronecker of two small dimensional weight matrices. The new cost function can be efficiently solved by combining Semi-Definite Programming (SDP) with Bi-Iterative Algorithm (BIA). The proposed algorithm has faster convergence rate, smaller training samples required and lower computational complexity comparable with that of AMA. The numerical examples are provided to demonstrate the effectiveness of the proposed algorithm.
2012, 34(4): 904-909.
doi: 10.3724/SP.J.1146.2011.00847
Abstract:
To solve the problem of Direction Of Departure-Direction Of Arrival (DOD-DOA) estimation in bistatic MIMO radar, a fast method based on joint matrix diagonalization is proposed. First, according to the structure of matched filter output, DOD-DOA estimation is transformed to joint matrix diagonalization problem using Singular Value Decomposition (SVD) and the theorem of rank-1 matrix determined. Then, the Single-Sweep Iterative (SSI) algorithm is used to solve it, and the transmit/receive steering matrices are obtained. Finally, the DOD-DOA can be estimated by spectrum analysis method. The proposed method utilizes all the information of matched filter output, avoiding two-dimensional spectrum peak searching, and possesses an accurate closed form solution at each iteration. The DOD and DOA are automatically paired. Compared with the existence approaches, the proposed algorithm gives better angle estimation accuracy, and the computational cost is effectively reduced. The effectiveness of the proposed method is demonstrated by simulation results.
To solve the problem of Direction Of Departure-Direction Of Arrival (DOD-DOA) estimation in bistatic MIMO radar, a fast method based on joint matrix diagonalization is proposed. First, according to the structure of matched filter output, DOD-DOA estimation is transformed to joint matrix diagonalization problem using Singular Value Decomposition (SVD) and the theorem of rank-1 matrix determined. Then, the Single-Sweep Iterative (SSI) algorithm is used to solve it, and the transmit/receive steering matrices are obtained. Finally, the DOD-DOA can be estimated by spectrum analysis method. The proposed method utilizes all the information of matched filter output, avoiding two-dimensional spectrum peak searching, and possesses an accurate closed form solution at each iteration. The DOD and DOA are automatically paired. Compared with the existence approaches, the proposed algorithm gives better angle estimation accuracy, and the computational cost is effectively reduced. The effectiveness of the proposed method is demonstrated by simulation results.
Analysis of Impact of Phase Center Variations in Linear Array Antena Downward-looking 3D-SAR Imaging
2012, 34(4): 910-916.
doi: 10.3724/SP.J.1146.2011.00776
Abstract:
The Linear Array Downward-Looking 3D-SAR (LADL 3D-SAR) achieves the ablity of resolving imaging in three dimensions using linear array antennas. The phase center variations of linear array antennas, which is unavoidable under real condition, will lead to echo phase errors that impact the imaging of 3D-SAR. In this paper, the analyzing model of the phase center variation is firstly established, then the impact of phase errors caused by phase center variations in LADL 3D-SAR is discussed based on stochastic process orthogonal expansion.The impact on sidlobe and the statistical relationship between ISLR and the deviation of phase center variations is derived analytically. Finally, the simulation results demonstrate the effectiveness of the theoretical analysis.
The Linear Array Downward-Looking 3D-SAR (LADL 3D-SAR) achieves the ablity of resolving imaging in three dimensions using linear array antennas. The phase center variations of linear array antennas, which is unavoidable under real condition, will lead to echo phase errors that impact the imaging of 3D-SAR. In this paper, the analyzing model of the phase center variation is firstly established, then the impact of phase errors caused by phase center variations in LADL 3D-SAR is discussed based on stochastic process orthogonal expansion.The impact on sidlobe and the statistical relationship between ISLR and the deviation of phase center variations is derived analytically. Finally, the simulation results demonstrate the effectiveness of the theoretical analysis.
2012, 34(4): 923-928.
doi: 10.3724/SP.J.1146.2011.00373
Abstract:
Small target detection in sea clutter is a challenging problem. This paper proposes a novel approach to solve it. Edge detection and improved Hough transform are applied to a set of range profiles, which are resulted from re-sampling and smoothing the range-time profiles in time direction. Using the two-step masks, the moving small target can be figured out in Hough domain. This approach has a wide application due to the independence on the model of sea clutter, and can be used to detect target without prior knowledge of the ocean and environment conditions. Real sea clutter data is used to illustrate the effectiveness and the practicability of the novel approach.
Small target detection in sea clutter is a challenging problem. This paper proposes a novel approach to solve it. Edge detection and improved Hough transform are applied to a set of range profiles, which are resulted from re-sampling and smoothing the range-time profiles in time direction. Using the two-step masks, the moving small target can be figured out in Hough domain. This approach has a wide application due to the independence on the model of sea clutter, and can be used to detect target without prior knowledge of the ocean and environment conditions. Real sea clutter data is used to illustrate the effectiveness and the practicability of the novel approach.
2012, 34(4): 929-935.
doi: 10.3724/SP.J.1146.2011.00856
Abstract:
This paper studies mainly the fractal property of the real sea clutter in frequency domain and the effects of different parameters to the fractal characteristics of the sea clutter spectrum. Firstly, this paper takes Fractional Brownian Motion (FBM) for example and interprets detailedly that the spectrum of the FBM is fractal on condition that the FBM time series is fractal. Then, X-band and S-band real sea clutter are used for the verification of the fractal property of the real sea clutter spectrum. At the same time, the effects of the different parameters are analyzed. The results show that the spectrum of the real sea clutter is self-similar in statistical meaning, namely it is fractal. Additionally, it is found that the fractal characteristics of the sea clutter and target echoes in frequency domain are different to some extent and this difference has the potential for weak target detection within sea clutter.
This paper studies mainly the fractal property of the real sea clutter in frequency domain and the effects of different parameters to the fractal characteristics of the sea clutter spectrum. Firstly, this paper takes Fractional Brownian Motion (FBM) for example and interprets detailedly that the spectrum of the FBM is fractal on condition that the FBM time series is fractal. Then, X-band and S-band real sea clutter are used for the verification of the fractal property of the real sea clutter spectrum. At the same time, the effects of the different parameters are analyzed. The results show that the spectrum of the real sea clutter is self-similar in statistical meaning, namely it is fractal. Additionally, it is found that the fractal characteristics of the sea clutter and target echoes in frequency domain are different to some extent and this difference has the potential for weak target detection within sea clutter.
2012, 34(4): 936-942.
doi: 10.3724/SP.J.1146.2011.00811
Abstract:
Space-Time Adaptive Processing (STAP) is an effective method for moving target detection in airborne radar. However, the Doppler frequency will change with time when the target has strong maneuvering, which would degrade the coherent integration performance of STAP significantly. For the detection and the parameter estimation of air maneuvering targets, a new algorithm is proposed in this paper, which combines STAP with FRactional Fourier Transform (FRFT). Space samples are used to reconstruct the time samples, which is equivalent to increasing the number of time samples within a Coherent Processing Interval (CPI). Consequently, the problem of the poor estimation performance by using the FRFT, which is caused by the limited number of time samples of airborne radar within a CPI, can be resolved. Numerical examples are provided to demonstrate the performance of the proposed algorithm.
Space-Time Adaptive Processing (STAP) is an effective method for moving target detection in airborne radar. However, the Doppler frequency will change with time when the target has strong maneuvering, which would degrade the coherent integration performance of STAP significantly. For the detection and the parameter estimation of air maneuvering targets, a new algorithm is proposed in this paper, which combines STAP with FRactional Fourier Transform (FRFT). Space samples are used to reconstruct the time samples, which is equivalent to increasing the number of time samples within a Coherent Processing Interval (CPI). Consequently, the problem of the poor estimation performance by using the FRFT, which is caused by the limited number of time samples of airborne radar within a CPI, can be resolved. Numerical examples are provided to demonstrate the performance of the proposed algorithm.
2012, 34(4): 943-949.
doi: 10.3724/SP.J.1146.2011.00720
Abstract:
Three dimensional SAR imaging technology based on linear array is one of the most important three dimensional SAR high resolution imaging methods. In this paper, transmit and receive antenna elements are placed to form cross-track thinned downward-looking array MIMO 3D-SAR. The imaging geometry and three dimensional echo signal model of cross-track thinned array MIMO 3D-SAR is analyzed, and an applicative imaging algorithm based on cross-track thinned array MIMO 3D-SAR is given. Finally, by emulation experiments, the imaging algorithm is verified and the 3D-SAR imaging results are analyzed.
Three dimensional SAR imaging technology based on linear array is one of the most important three dimensional SAR high resolution imaging methods. In this paper, transmit and receive antenna elements are placed to form cross-track thinned downward-looking array MIMO 3D-SAR. The imaging geometry and three dimensional echo signal model of cross-track thinned array MIMO 3D-SAR is analyzed, and an applicative imaging algorithm based on cross-track thinned array MIMO 3D-SAR is given. Finally, by emulation experiments, the imaging algorithm is verified and the 3D-SAR imaging results are analyzed.
2012, 34(4): 950-955.
doi: 10.3724/SP.J.1146.2011.00918
Abstract:
This paper proposes a new speckle reduction algorithm for Synthetic Aperture Radar (SAR) images. It is based on the Non Local (NL) means filter and improved by Structural SIMilarity (SSIM). Structure information is introduced into the despeckling method by measuring the similarity between small patches with SSIM. Some experiments on real SAR images, comparing with GammaMAP filter, Contourlet Hidden Markov Tree (CHMT) method, Bayes Least Squares-Gaussian Scale Mixtures (BLS-GSM) method and NL-means filter, demonstrate that the proposed algorithm is able to reduce efficiently speckle while retain edges and structures well.
This paper proposes a new speckle reduction algorithm for Synthetic Aperture Radar (SAR) images. It is based on the Non Local (NL) means filter and improved by Structural SIMilarity (SSIM). Structure information is introduced into the despeckling method by measuring the similarity between small patches with SSIM. Some experiments on real SAR images, comparing with GammaMAP filter, Contourlet Hidden Markov Tree (CHMT) method, Bayes Least Squares-Gaussian Scale Mixtures (BLS-GSM) method and NL-means filter, demonstrate that the proposed algorithm is able to reduce efficiently speckle while retain edges and structures well.
2012, 34(4): 956-962.
doi: 10.3724/SP.J.1146.2011.00699
Abstract:
To cope with the disadvantage that many receiving channels will be needed in traditional Long Baseline Interferometer (LBI) localization system, a novel emitter localization method using ambiguous phase difference measurements by a rotating LBI is proposed. High localization precision can be achieved in a short observing period with no localization ambiguity, and the amount of receiving channels can be reduced to two. To cope with the strong nonlinear problem caused by the phase difference ambiguity, a multiple hypothesis nonlinear least square localization algorithm is proposed. Flow of the algorithm is presented. The algorithm has some advantages, such as small computation burden and strong ability on resolving localization ambiguity. Moreover, localization precision of the algorithm can approach Cramer-Rao lower bound. Feasibility of the localization method and performance of the localization algorithm are validated through computer simulations.
To cope with the disadvantage that many receiving channels will be needed in traditional Long Baseline Interferometer (LBI) localization system, a novel emitter localization method using ambiguous phase difference measurements by a rotating LBI is proposed. High localization precision can be achieved in a short observing period with no localization ambiguity, and the amount of receiving channels can be reduced to two. To cope with the strong nonlinear problem caused by the phase difference ambiguity, a multiple hypothesis nonlinear least square localization algorithm is proposed. Flow of the algorithm is presented. The algorithm has some advantages, such as small computation burden and strong ability on resolving localization ambiguity. Moreover, localization precision of the algorithm can approach Cramer-Rao lower bound. Feasibility of the localization method and performance of the localization algorithm are validated through computer simulations.
2012, 34(4): 963-968.
doi: 10.3724/SP.J.1146.2011.0050
Abstract:
Based on analyzing the geometric model of SAR deception-jamming, a real-time deception-jamming algorithm for large scene is proposed. For real-time jamming, a large scene is divided into several sub-blocks and modulated separately. An analysis of jamming range is done according to the depth of focus in range and ambiguity of frequency in azimuth, which provides theroy foundation for blocking partition. By real-time adapting the focus-center of each sub-block, a good focusing image is obtained. Simulation results verify the effectiveness of this algorithm.
Based on analyzing the geometric model of SAR deception-jamming, a real-time deception-jamming algorithm for large scene is proposed. For real-time jamming, a large scene is divided into several sub-blocks and modulated separately. An analysis of jamming range is done according to the depth of focus in range and ambiguity of frequency in azimuth, which provides theroy foundation for blocking partition. By real-time adapting the focus-center of each sub-block, a good focusing image is obtained. Simulation results verify the effectiveness of this algorithm.
2012, 34(4): 969-975.
doi: 10.3724/SP.J.1146.2011.00698
Abstract:
Traditional Physical Optics (PO) is able to efficiently deal with 3D electromagnetic problem involving structures with electrically large dimension in the open space, usually obtaining calculation results with good precision. However, traditional PO appears to be inapposite for special complicated structures with electrically large scattering objects embedded between two parallel metal plates, so does the full wave method because of the large computation resource consumption. In this paper, a modified PO method based on Discrete Real Mirror Image theory (DRMI-PO) is proposed. The basic idea of DRMI-PO is to expand the electromagnetic current between the parallel metal plates into the Fourier series, and then linearly combine the field solution for each term, derived with the 2D PO. In comparison with Multilevel Fast Multipole Algorithm (MLFMA), DRMI-PO is able to keep enough precision and obtain higher efficiency when analyzing specific electromagnetic structures. As an application example, two terahertz fan-beam scanning antennas with different polarization are calculated with DRMI-PO and measured. The measured radiation patterns are in well agreement with the calculated results, which indicates the effectiveness of DRMI-PO.
Traditional Physical Optics (PO) is able to efficiently deal with 3D electromagnetic problem involving structures with electrically large dimension in the open space, usually obtaining calculation results with good precision. However, traditional PO appears to be inapposite for special complicated structures with electrically large scattering objects embedded between two parallel metal plates, so does the full wave method because of the large computation resource consumption. In this paper, a modified PO method based on Discrete Real Mirror Image theory (DRMI-PO) is proposed. The basic idea of DRMI-PO is to expand the electromagnetic current between the parallel metal plates into the Fourier series, and then linearly combine the field solution for each term, derived with the 2D PO. In comparison with Multilevel Fast Multipole Algorithm (MLFMA), DRMI-PO is able to keep enough precision and obtain higher efficiency when analyzing specific electromagnetic structures. As an application example, two terahertz fan-beam scanning antennas with different polarization are calculated with DRMI-PO and measured. The measured radiation patterns are in well agreement with the calculated results, which indicates the effectiveness of DRMI-PO.
2012, 34(4): 976-980.
doi: 10.3724/SP.J.1146.2011.00990
Abstract:
To reduce the noise figure in the tradition low noise amplifier, extend the bandwidth and achieve more precise step accuracy, a RF programmable amplifier with low noise single-ended differential circuit is proposed. The single-ended differential circuit uses the noise cancellation method to reduce the noise figure and uses the capacitance cross technique to extend the bandwidth. The improved source-level follower structure can achieve more precise step accuracy. The circuit is fabricated in 0.18mCMOS process. Under 1.8 V power supply and 170-870 MHz frequency signal input, the circuit achieves a 3.8 dB noise figure. The circuit shows a 55 dB gain control range by 0.8 dB each step. The overall power consumption is less than 14.76 mW, and the die area is 800m 600m . The test result shows that the circuit can provide lower noise figure and cover wider bandwidth while consumes the same current comparing with the tradition structure and the circuit can provide more precise step accuracy.
To reduce the noise figure in the tradition low noise amplifier, extend the bandwidth and achieve more precise step accuracy, a RF programmable amplifier with low noise single-ended differential circuit is proposed. The single-ended differential circuit uses the noise cancellation method to reduce the noise figure and uses the capacitance cross technique to extend the bandwidth. The improved source-level follower structure can achieve more precise step accuracy. The circuit is fabricated in 0.18mCMOS process. Under 1.8 V power supply and 170-870 MHz frequency signal input, the circuit achieves a 3.8 dB noise figure. The circuit shows a 55 dB gain control range by 0.8 dB each step. The overall power consumption is less than 14.76 mW, and the die area is 800m 600m . The test result shows that the circuit can provide lower noise figure and cover wider bandwidth while consumes the same current comparing with the tradition structure and the circuit can provide more precise step accuracy.
2012, 34(4): 986-991.
doi: 10.3724/SP.J.1146.2011.00915
Abstract:
Based on the property of the disjointed cubes that the logic operators OR and EXOR can replace each other, an algorithm of two level Mixed-Polarity Reed-Muller (MPRM) optimization is proposed. In the algorithm, by searching and decomposing the majority cubes of these disjointed cubes and replacing them with more compacted and less cubes, a minimized MPRM function is obtained. Further, an efficient approach for logic verification based on logic covers is also presented to check whether two functions are equal or not after logic minimization. The proposed algorithm is implemented in C and tested on MCNC benchmarks. Experimental results show that the proposed method can offer a compacted MPRM expression efficiently in contrast to the reported methods.
Based on the property of the disjointed cubes that the logic operators OR and EXOR can replace each other, an algorithm of two level Mixed-Polarity Reed-Muller (MPRM) optimization is proposed. In the algorithm, by searching and decomposing the majority cubes of these disjointed cubes and replacing them with more compacted and less cubes, a minimized MPRM function is obtained. Further, an efficient approach for logic verification based on logic covers is also presented to check whether two functions are equal or not after logic minimization. The proposed algorithm is implemented in C and tested on MCNC benchmarks. Experimental results show that the proposed method can offer a compacted MPRM expression efficiently in contrast to the reported methods.
2012, 34(4): 992-996.
doi: 10.3724/SP.J.1146.2011.00993
Abstract:
This paper presents a hierarchical cost aggregation-based fast stereo image matching method based on Webers law and guided filtering. Weber local descriptors for each color channel are firstly extracted from stereo pairs, and raw matching costs between the images are initialized by the descriptors. The matching costs are enhanced with guided filtering to extract the subsets of disparity candidates. Joint spatial discrete sampling and adaptive support weight are utilized to implement hierarchical cost aggregation on the candidate subsets. Then initial disparities from the subsets are selected fast and optimally. Modified bilateral filtering and symmetric warping-based post-processing are sequentially exploited in disparity refining to improve effectively ambiguous regions of initial disparity maps. The experimental results indicate that this proposed technique can obtain piecewise smooth, accurate and dense disparity map while eliminating effectively matching ambiguity. Being concise, fast and high efficiency, and it is robust to illumination change.
This paper presents a hierarchical cost aggregation-based fast stereo image matching method based on Webers law and guided filtering. Weber local descriptors for each color channel are firstly extracted from stereo pairs, and raw matching costs between the images are initialized by the descriptors. The matching costs are enhanced with guided filtering to extract the subsets of disparity candidates. Joint spatial discrete sampling and adaptive support weight are utilized to implement hierarchical cost aggregation on the candidate subsets. Then initial disparities from the subsets are selected fast and optimally. Modified bilateral filtering and symmetric warping-based post-processing are sequentially exploited in disparity refining to improve effectively ambiguous regions of initial disparity maps. The experimental results indicate that this proposed technique can obtain piecewise smooth, accurate and dense disparity map while eliminating effectively matching ambiguity. Being concise, fast and high efficiency, and it is robust to illumination change.
2012, 34(4): 997-1001.
doi: 10.3724/SP.J.1146.2011.00885
Abstract:
According to the problem of two-dimension (2-D) Direction Of Arrival (DOA) estimation accuracy low for wideband Linear Frequency Modulation (LFM) signals, a method of 2-D DOA estimation based on Radon-Wigner Transform (RWT) is proposed. In this method, by using the excellent time-frequency concentration performance of RWT which can eliminate cross interference terms and noise effectively in the background of multiple sources firstly, the number of target is determined by peak researching, and the array signal is reconstructed. Finally the MUSIC spatial spectrum analysis algorithm is used to estimate the 2-D DOA of multiple LFM signals. The simulation results show that the method based on RWT can efficiently estimate 2-D DOA for non-stationary signals.
According to the problem of two-dimension (2-D) Direction Of Arrival (DOA) estimation accuracy low for wideband Linear Frequency Modulation (LFM) signals, a method of 2-D DOA estimation based on Radon-Wigner Transform (RWT) is proposed. In this method, by using the excellent time-frequency concentration performance of RWT which can eliminate cross interference terms and noise effectively in the background of multiple sources firstly, the number of target is determined by peak researching, and the array signal is reconstructed. Finally the MUSIC spatial spectrum analysis algorithm is used to estimate the 2-D DOA of multiple LFM signals. The simulation results show that the method based on RWT can efficiently estimate 2-D DOA for non-stationary signals.
2012, 34(4): 1002-1006.
doi: 10.3724/SP.J.1146.2011.00957
Abstract:
This paper presents a new low power small area wide-band Voltage Controlled Oscillator (VCO) for Global Navigation Satellite System (GNSS) receiver. The VCO is separated in two discrete working regions from the characteristic of all-band GNSS signals. The power and phase noise can be optimized individually, the complexity of VCO is reduced and the area is saved. A technique of tuning curve linearization is used; The conventional VCO problem of having narrow effective tuning range of control voltage (VCTRL) is solved. A linear tuning curve in whole variation of VCTRL is kept, the Amplitude Modulation to Frequency Modulation (AM-FM) conversion is decreased, and the phase noise is allayed. The measured results show that the tuning range of frequency is 49.5%, the gain of VCO (KVCO) is constant when VCTRL varies from 0.1 to 0.9 V. Measured phase noise is lower than -120 dBc at 1 MHz offset, the entire VCO consumes 2 mA current, occupies 0.24 mm2 area. The proposed VCO is implemented in 0.13 m 1P6M process, and it is successfully applied to all-band GNSS receiver.
This paper presents a new low power small area wide-band Voltage Controlled Oscillator (VCO) for Global Navigation Satellite System (GNSS) receiver. The VCO is separated in two discrete working regions from the characteristic of all-band GNSS signals. The power and phase noise can be optimized individually, the complexity of VCO is reduced and the area is saved. A technique of tuning curve linearization is used; The conventional VCO problem of having narrow effective tuning range of control voltage (VCTRL) is solved. A linear tuning curve in whole variation of VCTRL is kept, the Amplitude Modulation to Frequency Modulation (AM-FM) conversion is decreased, and the phase noise is allayed. The measured results show that the tuning range of frequency is 49.5%, the gain of VCO (KVCO) is constant when VCTRL varies from 0.1 to 0.9 V. Measured phase noise is lower than -120 dBc at 1 MHz offset, the entire VCO consumes 2 mA current, occupies 0.24 mm2 area. The proposed VCO is implemented in 0.13 m 1P6M process, and it is successfully applied to all-band GNSS receiver.
2012, 34(4): 1007-1011.
doi: 10.3724/SP.J.1146.2011.00899
Abstract:
The mode switching is investigated for MIMO systems with considering four practical factors including channel estimation error, packet loss probability constraint, Automatic Repeat reQuest (ARQ) technique, discrete transmission rate and practical codec performance. The closed-form expressions of throughput are derived for spatial multiplexing and diversity modes, based on which the adaptive throughput-based mode switching is proposed. Numerical results show that the proposed scheme outperforms the fixed transmission mode schemes and the considered factors affect the throughput and the mode switching point.
The mode switching is investigated for MIMO systems with considering four practical factors including channel estimation error, packet loss probability constraint, Automatic Repeat reQuest (ARQ) technique, discrete transmission rate and practical codec performance. The closed-form expressions of throughput are derived for spatial multiplexing and diversity modes, based on which the adaptive throughput-based mode switching is proposed. Numerical results show that the proposed scheme outperforms the fixed transmission mode schemes and the considered factors affect the throughput and the mode switching point.
2012, 34(4): 1012-1016.
doi: 10.3724/SP.J.1146.2011.00884
Abstract:
A multi-threshold pipeline based on parallel completion is proposed to improve the throughput of asynchronous NULL Convention Logic (NCL) pipeline. With the special semi-static NCL threshold gates to be realized asynchronous combinational logic, data processing and completion detection of each pipeline stage are carried out parallelly, meanwhile, the data get through the pipeline by using serial mode. The series-parallel ways improve the throughput of the pipeline. Moreover, the static power of the pipeline in NULL cycle declines as well because of the new threshold gates. The proposed pipeline is simulated based on SMIC 0.18 m standard CMOS technology. Comparison results indicate that the throughput of the novel pipeline has an increment of 62.8% and the static power consumption is reduced by 40.5% with 4-bit NCL Ripper Adder serving as an asynchronous combinational logic. The proposed pipeline can be used to design high-speed low-power asynchronous circuit.
A multi-threshold pipeline based on parallel completion is proposed to improve the throughput of asynchronous NULL Convention Logic (NCL) pipeline. With the special semi-static NCL threshold gates to be realized asynchronous combinational logic, data processing and completion detection of each pipeline stage are carried out parallelly, meanwhile, the data get through the pipeline by using serial mode. The series-parallel ways improve the throughput of the pipeline. Moreover, the static power of the pipeline in NULL cycle declines as well because of the new threshold gates. The proposed pipeline is simulated based on SMIC 0.18 m standard CMOS technology. Comparison results indicate that the throughput of the novel pipeline has an increment of 62.8% and the static power consumption is reduced by 40.5% with 4-bit NCL Ripper Adder serving as an asynchronous combinational logic. The proposed pipeline can be used to design high-speed low-power asynchronous circuit.
2012, 34(4): 917-922.
doi: 10.3724/SP.J.1146.2011.00860
Abstract:
In order to improve the phase noise reduction effect and to avoid filtering useful phase jumps, this paper first investigates mode filtering algorithm and proposes a modified mode estimator based on shortest sub-interval seeking for interferometric phase. By further analysis of relationship between mode estimating parameter and phase resolution, an auto adaptive local phase center with variation of interferometric phase quality is constructed, based on which an adaptive median filtering method is presented. This method could overcome the edge blur problem caused by mode filter, and is also effective to avoid over filtering or under filtering caused by conventional spatial filter with uniform parameter. Meanwhile, it achieves better computation efficiency. Finally, experiments with simulated and real TerraSARX data are performed to testify the effectiveness of this method.
In order to improve the phase noise reduction effect and to avoid filtering useful phase jumps, this paper first investigates mode filtering algorithm and proposes a modified mode estimator based on shortest sub-interval seeking for interferometric phase. By further analysis of relationship between mode estimating parameter and phase resolution, an auto adaptive local phase center with variation of interferometric phase quality is constructed, based on which an adaptive median filtering method is presented. This method could overcome the edge blur problem caused by mode filter, and is also effective to avoid over filtering or under filtering caused by conventional spatial filter with uniform parameter. Meanwhile, it achieves better computation efficiency. Finally, experiments with simulated and real TerraSARX data are performed to testify the effectiveness of this method.
2012, 34(4): 981-985.
doi: 10.3724/SP.J.1146.2011.00382
Abstract:
Efficiency and linearity are two key features and design difficulties of Power Amplifier (PA). In this paper, a 2.5-2.6 GHz highly efficient GaN inverse class-F PA is designed. The input and output harmonic matching networks are designed through analytical approach. According to single tone test, the drain efficiency of the PA is over 75% at 2.55 GHz. In order to establish the behavioral model of the proposed PA and perform digital predistortion, the conventional Hammerstein model is modified so as the modeling accuracy is improved. For a 20 MHz 16-QAM OFDM signal with Peak-to-Average Power Ratio (PAPR) of 9.6 dB, the proposed PA is linearized by combining Crest Factor Reduction (CFR) and digital predistortion techniques. After linearization, Adjacent Channel Power Ratio (ACPR) of the PA is suppressed to below -48 dBc.
Efficiency and linearity are two key features and design difficulties of Power Amplifier (PA). In this paper, a 2.5-2.6 GHz highly efficient GaN inverse class-F PA is designed. The input and output harmonic matching networks are designed through analytical approach. According to single tone test, the drain efficiency of the PA is over 75% at 2.55 GHz. In order to establish the behavioral model of the proposed PA and perform digital predistortion, the conventional Hammerstein model is modified so as the modeling accuracy is improved. For a 20 MHz 16-QAM OFDM signal with Peak-to-Average Power Ratio (PAPR) of 9.6 dB, the proposed PA is linearized by combining Crest Factor Reduction (CFR) and digital predistortion techniques. After linearization, Adjacent Channel Power Ratio (ACPR) of the PA is suppressed to below -48 dBc.
