Email alert
2013 Vol. 35, No. 2
Display Method:
2013, 35(2): 255-260.
doi: 10.3724/SP.J.1146.2012.00862
Abstract:
Compressed Sensing (CS) technology paves the way for quick wideband spectrum detection and the WSN based on CS can provide Cognitive Radio (CR) users with the spectrum information. For the spectrum data detected in WSN, a two-dimensional compression Space-Frequency Compressed Sensing (SFCS) model is established in the space and frequency domain, and the corresponding algorithm for reconstruction is proposed and the performance of SFCS is analyzed. The simulation results show that SFCS needs less transmitted data than the traditional model on the same detection probability and the performance of Receiver Operating Characteristic (ROC) in the algorithm is better than that of traditional method on the same total compressed rate.
Compressed Sensing (CS) technology paves the way for quick wideband spectrum detection and the WSN based on CS can provide Cognitive Radio (CR) users with the spectrum information. For the spectrum data detected in WSN, a two-dimensional compression Space-Frequency Compressed Sensing (SFCS) model is established in the space and frequency domain, and the corresponding algorithm for reconstruction is proposed and the performance of SFCS is analyzed. The simulation results show that SFCS needs less transmitted data than the traditional model on the same detection probability and the performance of Receiver Operating Characteristic (ROC) in the algorithm is better than that of traditional method on the same total compressed rate.
2013, 35(2): 261-266.
doi: 10.3724/SP.J.1146.2012.00805
Abstract:
Because of the problem of the Minimum Mean Square Error with Successive Detection (MMSE-SD) algorithm with high computational complexity for OFDM systems in the presence of frequency and time double selective channels, an improved low-complexity MMSE-SD algorithm is proposed in this paper. The algorithm extends firstly the channel matrix and detection matrix, then establishes the relationship between them by using recursive inversion instead of direct inversion at each detection. Both theoretical analysis and simulation results show that the improved algorithm retains the performance of the original algorithm with much lower computational complexity compared to the original MMSE-SD algorithm; Compared to other algorithms, the improved algorithm balances the performance with the computational complexity. The larger normalization Doppler frequency shift, the better performance it possesses with the unchanged computational complexity.
Because of the problem of the Minimum Mean Square Error with Successive Detection (MMSE-SD) algorithm with high computational complexity for OFDM systems in the presence of frequency and time double selective channels, an improved low-complexity MMSE-SD algorithm is proposed in this paper. The algorithm extends firstly the channel matrix and detection matrix, then establishes the relationship between them by using recursive inversion instead of direct inversion at each detection. Both theoretical analysis and simulation results show that the improved algorithm retains the performance of the original algorithm with much lower computational complexity compared to the original MMSE-SD algorithm; Compared to other algorithms, the improved algorithm balances the performance with the computational complexity. The larger normalization Doppler frequency shift, the better performance it possesses with the unchanged computational complexity.
2013, 35(2): 267-272.
doi: 10.3724/SP.J.1146.2012.00787
Abstract:
Dynamic channel sharing policy is the key technology to improve the spectrum utilization in cognitive radio networks. First, the scenario is considered that several cognitive networks with finite user population coexist. Then, a three-dimensional Markov chain model is developed to analyze the performance of the partial channel sharing policy in the cases with and without channel handoff. Finally, the blocking probability, the forced termination probability, handoff probability of the users and the system throughput are analyzed by comparing with the static spectrum allocation policy and the hierarchical sharing policy. The simulation results show that the system with partial channel sharing policy will gain higher throughput by tolerating lower handoff probability and forced termination probability.
Dynamic channel sharing policy is the key technology to improve the spectrum utilization in cognitive radio networks. First, the scenario is considered that several cognitive networks with finite user population coexist. Then, a three-dimensional Markov chain model is developed to analyze the performance of the partial channel sharing policy in the cases with and without channel handoff. Finally, the blocking probability, the forced termination probability, handoff probability of the users and the system throughput are analyzed by comparing with the static spectrum allocation policy and the hierarchical sharing policy. The simulation results show that the system with partial channel sharing policy will gain higher throughput by tolerating lower handoff probability and forced termination probability.
2013, 35(2): 273-279.
doi: 10.3724/SP.J.1146.2012.00609
Abstract:
An improved multimodulus blind equalization algorithm for QAM signal is presented. The real part (imaginary part) information is used to construct cost function. Moreover, conjugate gradient algorithm is proposed to optimize the equalizer coefficients. With its equalization performance comparable to the conventional Constant Modulus Algorithm (CMA) and MultiModulus Algorithm (MMA), the proposed algorithm is quadratic convergent and has the advantages of better convergent performance and less computational complexity. Finally, the reliability and validity of this algorithm are analyzed by Bit Error Rate (BER) and rate of convergence. Its good performance is verified by simulation results.
An improved multimodulus blind equalization algorithm for QAM signal is presented. The real part (imaginary part) information is used to construct cost function. Moreover, conjugate gradient algorithm is proposed to optimize the equalizer coefficients. With its equalization performance comparable to the conventional Constant Modulus Algorithm (CMA) and MultiModulus Algorithm (MMA), the proposed algorithm is quadratic convergent and has the advantages of better convergent performance and less computational complexity. Finally, the reliability and validity of this algorithm are analyzed by Bit Error Rate (BER) and rate of convergence. Its good performance is verified by simulation results.
2013, 35(2): 280-284.
doi: 10.3724/SP.J.1146.2012.01186
Abstract:
Considering the issue of In-phase and Quadrature-phase (IQ) imbalances in MIMO-OFDM systems with sparse multipath channels, a method jointing IQ imbalance with sparse channel estimation is proposed. The time-domain model for joint IQ imbalance and sparse channel estimation is derived and an estimation scheme based onl1l2 norm optimization model and Parallel Coordinate Descent (PCD) algorithm is proposed. Simulation results show that the proposed scheme improves obviously the system performance compared with the traditional frequency-domain Least Squares (LS) and Matching Pursuit (MP) algorithms.
Considering the issue of In-phase and Quadrature-phase (IQ) imbalances in MIMO-OFDM systems with sparse multipath channels, a method jointing IQ imbalance with sparse channel estimation is proposed. The time-domain model for joint IQ imbalance and sparse channel estimation is derived and an estimation scheme based onl1l2 norm optimization model and Parallel Coordinate Descent (PCD) algorithm is proposed. Simulation results show that the proposed scheme improves obviously the system performance compared with the traditional frequency-domain Least Squares (LS) and Matching Pursuit (MP) algorithms.
2013, 35(2): 285-290.
doi: 10.3724/SP.J.1146.2012.00813
Abstract:
Introducing relay to wireless networks results in severer energy consumption. The circuit power consumption of devices can not be neglected in energy efficient problem. In this paper, Orthogonal Frequency Division Multiplexing (OFDM)-based frequency-selective channel is considered in Decode-and-Forward (DF) relay link. An optimal allocation method is proposed, in which the transmit power in each hop is adaptively allocated to maximize the energy efficiency. The effects of the rate and power constraints on energy efficiency are also analyzed. Simulation results show that the proposed energy efficient power allocation method can achieve the highest energy efficiency while ensuring the high data rate.
Introducing relay to wireless networks results in severer energy consumption. The circuit power consumption of devices can not be neglected in energy efficient problem. In this paper, Orthogonal Frequency Division Multiplexing (OFDM)-based frequency-selective channel is considered in Decode-and-Forward (DF) relay link. An optimal allocation method is proposed, in which the transmit power in each hop is adaptively allocated to maximize the energy efficiency. The effects of the rate and power constraints on energy efficiency are also analyzed. Simulation results show that the proposed energy efficient power allocation method can achieve the highest energy efficiency while ensuring the high data rate.
2013, 35(2): 291-296.
doi: 10.3724/SP.J.1146.2012.01085
Abstract:
Distributed power control algorithms for systems with hard SINR (Signal-to-Interference plus Noise Satio) constraints may diverge when infeasibility arises. Non-cooperative game-theoretic approach is applied to model the power control of the transmitters in 60 GHz millimeter wave communication systems. Combining with the feature of 60 GHz wireless network, a new Unitary Step Utility-based Transmit Power Control (USUTPC) is proposed to improve the performance. A detailed analysis for the existence and uniqueness of Nash equilibrium for the above non-cooperative game is presented. The simulation results denote that the USUTPC is quiet simple and reasonable. Meanwhile, the influence of different smooth parameter and price factor on simulation result is analyzed.
Distributed power control algorithms for systems with hard SINR (Signal-to-Interference plus Noise Satio) constraints may diverge when infeasibility arises. Non-cooperative game-theoretic approach is applied to model the power control of the transmitters in 60 GHz millimeter wave communication systems. Combining with the feature of 60 GHz wireless network, a new Unitary Step Utility-based Transmit Power Control (USUTPC) is proposed to improve the performance. A detailed analysis for the existence and uniqueness of Nash equilibrium for the above non-cooperative game is presented. The simulation results denote that the USUTPC is quiet simple and reasonable. Meanwhile, the influence of different smooth parameter and price factor on simulation result is analyzed.
2013, 35(2): 297-302.
doi: 10.3724/SP.J.1146.2012.00403
Abstract:
The construction of non-binary quasi-cyclic low-density parity-check codes are investigated. Important factors that influence the performance of non-binary LDPC codes are discussed in detail. The circle length and connectivity are taken into account in the construction. The selection of shift values to extend the mother matrix and the replacement of non-zero elements over GF(q) for the parity check matrix are studied. Meanwhile, a suboptimum method with lower complexity is proposed to search for the solution according to the new formula. The method is applied to the construction of non-binary LDPC codes over different Galois fields. Simulation results show that the codes constructed by this method outperform the corresponding binary codes by 0.2 dB at FER around10-4 on the BPSK AWGN channel. The performance gap is larger when the order of the Galois field equals the order of the modulation. Compared with non-binary cycle code with the same code rate and approximate code length, the constructed non-binary LDPC code gets an improvement of 0.25 dB at FER around 10-4.
The construction of non-binary quasi-cyclic low-density parity-check codes are investigated. Important factors that influence the performance of non-binary LDPC codes are discussed in detail. The circle length and connectivity are taken into account in the construction. The selection of shift values to extend the mother matrix and the replacement of non-zero elements over GF(q) for the parity check matrix are studied. Meanwhile, a suboptimum method with lower complexity is proposed to search for the solution according to the new formula. The method is applied to the construction of non-binary LDPC codes over different Galois fields. Simulation results show that the codes constructed by this method outperform the corresponding binary codes by 0.2 dB at FER around10-4 on the BPSK AWGN channel. The performance gap is larger when the order of the Galois field equals the order of the modulation. Compared with non-binary cycle code with the same code rate and approximate code length, the constructed non-binary LDPC code gets an improvement of 0.25 dB at FER around 10-4.
2013, 35(2): 303-309.
doi: 10.3724/SP.J.1146.2012.00725
Abstract:
An attack on f9 algorithm based on 4-round KASUMI is given with a single-key method. The meet-in -the-middle thought is applied into the attack and the based key set and the exhaustive key set are chosen. Then f9 algorithm is attacked with the linear relationship between plaintext and K3. At the same time, collision properties and table-lookups are applied to reduce the time complexity. As a result, with 232 plaintexts with the corresponding MACs, the attack needs 2125.85 f9 calculations with 236 memory to recovery all the key.
An attack on f9 algorithm based on 4-round KASUMI is given with a single-key method. The meet-in -the-middle thought is applied into the attack and the based key set and the exhaustive key set are chosen. Then f9 algorithm is attacked with the linear relationship between plaintext and K3. At the same time, collision properties and table-lookups are applied to reduce the time complexity. As a result, with 232 plaintexts with the corresponding MACs, the attack needs 2125.85 f9 calculations with 236 memory to recovery all the key.
2013, 35(2): 310-313.
doi: 10.3724/SP.J.1146.2012.00629
Abstract:
In order to solve the data integrality issue in the cloud computing, a new interactive integrality checking scheme is proposed. This scheme firstly constructs a perfect cartesian message authentication function type I and its equivalent function, then use synchronous verification value and the equivalent function to determine response message authentication value to detect integrity. Analysis results show that this new scheme can correctly check cloud file,s integrality with the difficulty of factoring large integer and only request constant computation, storage and constant network resource for the users, which is more computationally efficient than others.
In order to solve the data integrality issue in the cloud computing, a new interactive integrality checking scheme is proposed. This scheme firstly constructs a perfect cartesian message authentication function type I and its equivalent function, then use synchronous verification value and the equivalent function to determine response message authentication value to detect integrity. Analysis results show that this new scheme can correctly check cloud file,s integrality with the difficulty of factoring large integer and only request constant computation, storage and constant network resource for the users, which is more computationally efficient than others.
2013, 35(2): 314-319.
doi: 10.3724/SP.J.1146.2012.00984
Abstract:
The restriction of the network performance such as the processing delay of control plane, the utilization probability and the burst collision probability on the assembling algorithm is analyzed and studied. The results verify that the data plane is more sensitive than the control plane with the assembling algorithm of edge nodes. The network with core nodes without wavelength converter has no valid assembling threshold under the multiple objective restrictions. When implemented with a general optical exchange connector and under the desired burst collision ratio 10-4, the core node needs at least thirty wavelength converters per output fiber link to satisfy the performance requirement.
The restriction of the network performance such as the processing delay of control plane, the utilization probability and the burst collision probability on the assembling algorithm is analyzed and studied. The results verify that the data plane is more sensitive than the control plane with the assembling algorithm of edge nodes. The network with core nodes without wavelength converter has no valid assembling threshold under the multiple objective restrictions. When implemented with a general optical exchange connector and under the desired burst collision ratio 10-4, the core node needs at least thirty wavelength converters per output fiber link to satisfy the performance requirement.
2013, 35(2): 320-327.
doi: 10.3724/SP.J.1146.2012.00795
Abstract:
Because the resources in cognitive wireless networks are high dynamic, there must be a special transmission control mechanism to coordinate the transmission of each node. A novel backpressure transmission control mechanism is proposed to supply an efficient and stable end to end transmission. The novel control mechanism takes full advantage of the adjustment ability of the local nodes to solve the undesirable network conditions in a local area and make the source node making the end to end control measures softly. An extensive simulations show that the transmission control mechanism is suitable for the dynamic cognitive wireless networks, and is capable of making the end to end transmission efficient and stable.
Because the resources in cognitive wireless networks are high dynamic, there must be a special transmission control mechanism to coordinate the transmission of each node. A novel backpressure transmission control mechanism is proposed to supply an efficient and stable end to end transmission. The novel control mechanism takes full advantage of the adjustment ability of the local nodes to solve the undesirable network conditions in a local area and make the source node making the end to end control measures softly. An extensive simulations show that the transmission control mechanism is suitable for the dynamic cognitive wireless networks, and is capable of making the end to end transmission efficient and stable.
2013, 35(2): 328-334.
doi: 10.3724/SP.J.1146.2012.00723
Abstract:
In order to remove the vast information redundancy in dense Wireless Sensor Networks (WSN), a distributed source coding algorithm for clustering WSN is proposed. The algorithm uses the correlation among the sources to define the coding sequence and their reference source with the side information as the initial reference source and each source codes its data with respect to its reference source, and the receiver decodes the data according to the coding sequence and reference source. In addition, a decoding algorithm with low complexity for modulus coding is given. The analysis and simulation results show that the proposed algorithm applied to clustering routing protocol can effectively reduce the number of bits to be sent, and thereby decrease the energy consumption of network and prolong the network lifetime.
In order to remove the vast information redundancy in dense Wireless Sensor Networks (WSN), a distributed source coding algorithm for clustering WSN is proposed. The algorithm uses the correlation among the sources to define the coding sequence and their reference source with the side information as the initial reference source and each source codes its data with respect to its reference source, and the receiver decodes the data according to the coding sequence and reference source. In addition, a decoding algorithm with low complexity for modulus coding is given. The analysis and simulation results show that the proposed algorithm applied to clustering routing protocol can effectively reduce the number of bits to be sent, and thereby decrease the energy consumption of network and prolong the network lifetime.
2013, 35(2): 335-340.
doi: 10.3724/SP.J.1146.2012.00854
Abstract:
An instruction set extension for Counter mode with Cipher block chaining Message authentication code protocol using Advanced Encryption Standard algorithm (AES-CCM) protocol in IEEE802.15.4 is presented based on AES fast algorithms and NiosII processor. The logical relationship between the lookup table used for round transformation and S-box is derived, then the S-box value is calculated with composite field transform method in hardware circuit, thereby eliminated the consumption of on-chip memory. The scheme is verified on EP2C35 chip, and the design and experimental data of look-up table method of the instruction set extension design and co-processor are also proposed for compare. This schemes increases the speedup by 174.6 times than software implementation, only uses 223 logic elements as 9.5% of coprocessor, throughput achieves 668.7 kbps, and reduces significantly the cost and power consumption of wireless sensor network node equipments.
An instruction set extension for Counter mode with Cipher block chaining Message authentication code protocol using Advanced Encryption Standard algorithm (AES-CCM) protocol in IEEE802.15.4 is presented based on AES fast algorithms and NiosII processor. The logical relationship between the lookup table used for round transformation and S-box is derived, then the S-box value is calculated with composite field transform method in hardware circuit, thereby eliminated the consumption of on-chip memory. The scheme is verified on EP2C35 chip, and the design and experimental data of look-up table method of the instruction set extension design and co-processor are also proposed for compare. This schemes increases the speedup by 174.6 times than software implementation, only uses 223 logic elements as 9.5% of coprocessor, throughput achieves 668.7 kbps, and reduces significantly the cost and power consumption of wireless sensor network node equipments.
2013, 35(2): 341-348.
doi: 10.3724/SP.J.1146.2012.00654
Abstract:
A low latency router architecture based Pre-Allocated Path Router (PAPR) is proposed for the Network-on-Chip (NoC). The advanced-routing algorithm and pre-allocated path are implemented to reduce the depth of the pipeline. The pre-allocated virtual channel arbitration is guaranteed by the advanced-routing. The packet transmission time in the network will not be affected even if the pre-allocation of virtual channel arbitration fails. The arbitration algorithm BSTS (Buffer Status ) takes the buffer status of the current node and downstream node into consideration. The new router not only reduces the packet waiting time in the buffer, but also lowers the probabilities of the idle buffer. According to the simulation results, compared with Generic Virtual Channel Router with iSLIP (GVCR-iSLIP) (iterative Round-Robin Matching with Serial Line Interface Protocal (SLIP)), the PAPR reduces the End To End (ETE) delay by 24.5% and improves the throughput by 27.5%; Compared with GVCR with Round-Robin Matching (RRM) (GVCR-RRM), the ETE delay is decreased by 39.2% and the throughput is increased by 47.2%. The router logic unit overhead and the power consumption are only increased by 8.9% and 5.9% respectively.
A low latency router architecture based Pre-Allocated Path Router (PAPR) is proposed for the Network-on-Chip (NoC). The advanced-routing algorithm and pre-allocated path are implemented to reduce the depth of the pipeline. The pre-allocated virtual channel arbitration is guaranteed by the advanced-routing. The packet transmission time in the network will not be affected even if the pre-allocation of virtual channel arbitration fails. The arbitration algorithm BSTS (Buffer Status ) takes the buffer status of the current node and downstream node into consideration. The new router not only reduces the packet waiting time in the buffer, but also lowers the probabilities of the idle buffer. According to the simulation results, compared with Generic Virtual Channel Router with iSLIP (GVCR-iSLIP) (iterative Round-Robin Matching with Serial Line Interface Protocal (SLIP)), the PAPR reduces the End To End (ETE) delay by 24.5% and improves the throughput by 27.5%; Compared with GVCR with Round-Robin Matching (RRM) (GVCR-RRM), the ETE delay is decreased by 39.2% and the throughput is increased by 47.2%. The router logic unit overhead and the power consumption are only increased by 8.9% and 5.9% respectively.
2013, 35(2): 349-354.
doi: 10.3724/SP.J.1146.2012.00946
Abstract:
Current Quality of Service (QoS)-based service selection approaches pay little attention to personal properties and characteristics of service requesters. However, Collaborative Filtering (CF)-based service selection approaches fail to consider the trustworthiness of recommenders and can not resist malicious feedback from recommenders. This paper introduces user correlation to embody the impact of personal characteristics of service requesters on selection process, computes creditability of recommendation, and employs analytic hierarchy process to decide the weight of each factor in service reputation. With the integration of trust evaluation methods and CF techniques, a Trustworthy Services Selection Model based on Collaborative Filtering (TSSMCF) is presented. Simulation experiments demonstrate that TSSMCF model can not only improve the efficiency of service selection but also effectively avoid malicious attack from service recommenders.
Current Quality of Service (QoS)-based service selection approaches pay little attention to personal properties and characteristics of service requesters. However, Collaborative Filtering (CF)-based service selection approaches fail to consider the trustworthiness of recommenders and can not resist malicious feedback from recommenders. This paper introduces user correlation to embody the impact of personal characteristics of service requesters on selection process, computes creditability of recommendation, and employs analytic hierarchy process to decide the weight of each factor in service reputation. With the integration of trust evaluation methods and CF techniques, a Trustworthy Services Selection Model based on Collaborative Filtering (TSSMCF) is presented. Simulation experiments demonstrate that TSSMCF model can not only improve the efficiency of service selection but also effectively avoid malicious attack from service recommenders.
2013, 35(2): 355-360.
doi: 10.3724/SP.J.1146.2012.00711
Abstract:
As mathematical fundamental of stochastic computing system, transfer function of variance and expected value based on Bernoulli distribution is not accurate and general in system analysis. A novel mathematic method, hypergeometric decomposition is proposed to solve this problem; it offers a general way to calculate transfer function of expected value and variance under more complicated circumstance. There are four groups of transfer function proposed here, which proves the effectiveness of stochastic computing system in a more general way; also they offer a better way to evaluate stochastic system. Compared with traditional bit-level simulation, evaluation method based on variance is time saving, accurate and comprehensive. New variance transfer function includes type of input random stream into performance analysis for the first time, which proves that specific length of stochastic sequence can maximize system performance.
As mathematical fundamental of stochastic computing system, transfer function of variance and expected value based on Bernoulli distribution is not accurate and general in system analysis. A novel mathematic method, hypergeometric decomposition is proposed to solve this problem; it offers a general way to calculate transfer function of expected value and variance under more complicated circumstance. There are four groups of transfer function proposed here, which proves the effectiveness of stochastic computing system in a more general way; also they offer a better way to evaluate stochastic system. Compared with traditional bit-level simulation, evaluation method based on variance is time saving, accurate and comprehensive. New variance transfer function includes type of input random stream into performance analysis for the first time, which proves that specific length of stochastic sequence can maximize system performance.
2013, 35(2): 361-367.
doi: 10.3724/SP.J.1146.2012.00790
Abstract:
A novel hybrid multi-valued Discrete Particle Swarm Optimization (DPSO) algorithm for Mixed- Polarity Reed-Muller (MPRM) minimization of Boolean function system is proposed. To solve the problem of diversity loss, improve the optimized result and balance the efficiency and precision of DPSO, multi-swarm cooperative optimization is employed, and three update and mutation strategies of update with probabilistic mutation, update with no duplicates and mutation with best duplicates between swarms are proposed. The experimental results show that compared with Simulated Annealing Genetic Algorithm (SAGA), the proposed algorithm can obtain similar optimized results and improve the time efficiency of MPRM minimization.
A novel hybrid multi-valued Discrete Particle Swarm Optimization (DPSO) algorithm for Mixed- Polarity Reed-Muller (MPRM) minimization of Boolean function system is proposed. To solve the problem of diversity loss, improve the optimized result and balance the efficiency and precision of DPSO, multi-swarm cooperative optimization is employed, and three update and mutation strategies of update with probabilistic mutation, update with no duplicates and mutation with best duplicates between swarms are proposed. The experimental results show that compared with Simulated Annealing Genetic Algorithm (SAGA), the proposed algorithm can obtain similar optimized results and improve the time efficiency of MPRM minimization.
2013, 35(2): 368-375.
doi: 10.3724/SP.J.1146.2012.00897
Abstract:
An analytic solution to the Non-Negative Eigenvalue Decomposition (NNED) in the non-reflection symmetry case is derived for the first time, which is named as NNED with non-reflection symmetry. It is applied to the Freeman decomposition, and then a Freeman decomposition based on NNED with non-reflection symmetry is proposed. During the Freeman decomposition, the NNED with non-reflection symmetry is used to extract volume scattering power, and adjust volume scattering power, double-bounce scattering power and surface scattering power to ensure the remainder covariance matrix has no negative eigenvalues. Compared with the Freeman decomposition based on NNED with reflection symmetry, the proposed decomposition method can availably use the non-diagonal elements which are regarded as zeros in the reflection symmetry case, and can ensure the remainder covariance matrix has no negative eigenvalues. The real-POLSAR-data experiment shows the proposed decomposition method can markedly enhance the double-bounce scattering power and weaken the volume scattering power.
An analytic solution to the Non-Negative Eigenvalue Decomposition (NNED) in the non-reflection symmetry case is derived for the first time, which is named as NNED with non-reflection symmetry. It is applied to the Freeman decomposition, and then a Freeman decomposition based on NNED with non-reflection symmetry is proposed. During the Freeman decomposition, the NNED with non-reflection symmetry is used to extract volume scattering power, and adjust volume scattering power, double-bounce scattering power and surface scattering power to ensure the remainder covariance matrix has no negative eigenvalues. Compared with the Freeman decomposition based on NNED with reflection symmetry, the proposed decomposition method can availably use the non-diagonal elements which are regarded as zeros in the reflection symmetry case, and can ensure the remainder covariance matrix has no negative eigenvalues. The real-POLSAR-data experiment shows the proposed decomposition method can markedly enhance the double-bounce scattering power and weaken the volume scattering power.
2013, 35(2): 376-380.
doi: 10.3724/SP.J.1146.2012.00840
Abstract:
The problem of detecting a weak target in dynamic clutter scenarios is analyzed with a polarimetric high-resolution radar. The heavy-tailed clutter is modeled by the compound-Gaussian process with inverse Gamma distributed texture. With training data to estimate covariance matrix of clutter, an adaptive polarimetric detector based on generalized likelihood ratio test criterion is presented for this heavy-tailed compound-Gaussian clutter. Then, the analytic expression of false alarm is derived to prove its constant false alarm rate property with respect to the clutter covariance matrix. The simulation results confirm the effectiveness of the proposed detector.
The problem of detecting a weak target in dynamic clutter scenarios is analyzed with a polarimetric high-resolution radar. The heavy-tailed clutter is modeled by the compound-Gaussian process with inverse Gamma distributed texture. With training data to estimate covariance matrix of clutter, an adaptive polarimetric detector based on generalized likelihood ratio test criterion is presented for this heavy-tailed compound-Gaussian clutter. Then, the analytic expression of false alarm is derived to prove its constant false alarm rate property with respect to the clutter covariance matrix. The simulation results confirm the effectiveness of the proposed detector.
2013, 35(2): 381-387.
doi: 10.3724/SP.J.1146.2012.00843
Abstract:
Detecting high-speed multi-target is affected by range migration, Doppler spread and velocity ambiguity in the linear frequency modulation pulse compression radar. In order to solve these problems, the joint frequency-domain scaling pulse compression processing and Lvs Method (LvM, 2011) is utilized to realize coherent integration of target signal firstly. And then, parametric detection for high-speed multi-target is completed with Doppler frequency ambiguity searching based method. The computational load of target parametric detection is reduced by the proposed frequency-domain scaling pulse compression processing, which can compensate range migration and Doppler frequency ambiguity simultaneously. The performance of the target detection and moving parameters estimation is improved owing to the utilization of the coherent integration in low signal-to-noise ratio. With the computational load analysis, computer simulation and raw radar data results, the validity of the proposed algorithm is demonstrated.
Detecting high-speed multi-target is affected by range migration, Doppler spread and velocity ambiguity in the linear frequency modulation pulse compression radar. In order to solve these problems, the joint frequency-domain scaling pulse compression processing and Lvs Method (LvM, 2011) is utilized to realize coherent integration of target signal firstly. And then, parametric detection for high-speed multi-target is completed with Doppler frequency ambiguity searching based method. The computational load of target parametric detection is reduced by the proposed frequency-domain scaling pulse compression processing, which can compensate range migration and Doppler frequency ambiguity simultaneously. The performance of the target detection and moving parameters estimation is improved owing to the utilization of the coherent integration in low signal-to-noise ratio. With the computational load analysis, computer simulation and raw radar data results, the validity of the proposed algorithm is demonstrated.
2013, 35(2): 388-393.
doi: 10.3724/SP.J.1146.2012.00238
Abstract:
The sparse signal reconstruction with Compressed Sensing (CS) is actually solving a system of underdetermined linear equations within the signal sparsity, of which one focus is to reduce recovery errors by the type of iteratively weightedLp(001,p=2) algorithms recently. The Basis Pursuit-Moore-Penrose Inverse Matrix (BP-MPIM) algorithm is proposed in this paper. First, nonzero element coordinates of the sparse signal are acquired by the basis pursuit algorithm, which are renamed with the sparse signal support set (corresponding with columns of the measure matrix). Then, the sparse signal recovery is solved from a set of superdetermined linear equations, which is composed of the submatrix of the sampling matrix and compressed sensing measurements. At the same time, it is proved that the reconstruction of sparse signals by this new algorithm is the one and only minimize L2 norm. Both simulative sparse signals and pulse compressed data of wideband radar echoes indicate that the new algorithm has less recovery errors than the previous algorithms, which are just in the support set.
The sparse signal reconstruction with Compressed Sensing (CS) is actually solving a system of underdetermined linear equations within the signal sparsity, of which one focus is to reduce recovery errors by the type of iteratively weightedLp(001,p=2) algorithms recently. The Basis Pursuit-Moore-Penrose Inverse Matrix (BP-MPIM) algorithm is proposed in this paper. First, nonzero element coordinates of the sparse signal are acquired by the basis pursuit algorithm, which are renamed with the sparse signal support set (corresponding with columns of the measure matrix). Then, the sparse signal recovery is solved from a set of superdetermined linear equations, which is composed of the submatrix of the sampling matrix and compressed sensing measurements. At the same time, it is proved that the reconstruction of sparse signals by this new algorithm is the one and only minimize L2 norm. Both simulative sparse signals and pulse compressed data of wideband radar echoes indicate that the new algorithm has less recovery errors than the previous algorithms, which are just in the support set.
2013, 35(2): 394-400.
doi: 10.3724/SP.J.1146.2012.00920
Abstract:
Due to the ambiguities in Direction Of Arrival (DOA) estimation for an interferometric array VHF radar, a novel ambiguity-resolution algorithm based on matrix completion is proposed. The algorithm with additional elements for ambiguity resolution in interferometry is extended to the interferometric array. Then the interferometric array becomes a completion array or fully augmentable array. The well-known Direct Augmentation Approach (DAA), MUSIC (MUltiple SIgnal Classification) and association method are orderly utilized to resolve the ambiguities and obtain nonambiguous but low-variance DOA estimation. An interferometric spatial smoothing technique for decorrelating coherent signals is also proposed in terms of the array geometry. Simulation results and the results of real data demonstrate the validity and effectiveness of the proposed algorithms. The matrix-completion algorithm has some computational and real-time advantages at the cost of a small increase in hardware.
Due to the ambiguities in Direction Of Arrival (DOA) estimation for an interferometric array VHF radar, a novel ambiguity-resolution algorithm based on matrix completion is proposed. The algorithm with additional elements for ambiguity resolution in interferometry is extended to the interferometric array. Then the interferometric array becomes a completion array or fully augmentable array. The well-known Direct Augmentation Approach (DAA), MUSIC (MUltiple SIgnal Classification) and association method are orderly utilized to resolve the ambiguities and obtain nonambiguous but low-variance DOA estimation. An interferometric spatial smoothing technique for decorrelating coherent signals is also proposed in terms of the array geometry. Simulation results and the results of real data demonstrate the validity and effectiveness of the proposed algorithms. The matrix-completion algorithm has some computational and real-time advantages at the cost of a small increase in hardware.
2013, 35(2): 401-405.
doi: 10.3724/SP.J.1146.2012.00846
Abstract:
Skywave over-the-horizon radar usually detects targets over the sea surface using Range-Doppler (RD) processing. RD processing can obtain the slant range and Doppler frequency of the target, but the altitude can not be derived. An algorithm based on the matched-field processing is proposed to estimate the targets altitude. The algorithm uses the return signal in a single dwell. Simulation result shows that under the specified simulation scenario, the altitude estimation error is within 2000 m. Comparing with the log likelihood function algorithm, the proposed algorithm has better robustness.
Skywave over-the-horizon radar usually detects targets over the sea surface using Range-Doppler (RD) processing. RD processing can obtain the slant range and Doppler frequency of the target, but the altitude can not be derived. An algorithm based on the matched-field processing is proposed to estimate the targets altitude. The algorithm uses the return signal in a single dwell. Simulation result shows that under the specified simulation scenario, the altitude estimation error is within 2000 m. Comparing with the log likelihood function algorithm, the proposed algorithm has better robustness.
2013, 35(2): 406-412.
doi: 10.3724/SP.J.1146.2012.00567
Abstract:
Integration of radar and communication on the electronic war platform is an effective method to reduce their volume, electromagnetic interference. In allusion to mutual interference and incompatible issue between suitable wareforms design of integrated radar and communication, following the principle of signal sharing, integration of radar and communication system and the correspondence processing scheme is presented based on FM orthogonal multicarrier chirp signal. First, based on wideband ambiguity function, the characteristics of the multicarrier integrated signal are analyzed in detail. The system performance and signal processing of integrated signal are analyzed. Simulation results and theoretical analysis show that the integrated signal can satisfy conventional radar detection and have low bit error rate under 20% rate of spectrum overlapping.
Integration of radar and communication on the electronic war platform is an effective method to reduce their volume, electromagnetic interference. In allusion to mutual interference and incompatible issue between suitable wareforms design of integrated radar and communication, following the principle of signal sharing, integration of radar and communication system and the correspondence processing scheme is presented based on FM orthogonal multicarrier chirp signal. First, based on wideband ambiguity function, the characteristics of the multicarrier integrated signal are analyzed in detail. The system performance and signal processing of integrated signal are analyzed. Simulation results and theoretical analysis show that the integrated signal can satisfy conventional radar detection and have low bit error rate under 20% rate of spectrum overlapping.
2013, 35(2): 413-418.
doi: 10.3724/SP.J.1146.2012.01060
Abstract:
In order to analyze the micro-movement component of the ballistic targets at the middle trajectory, the accurate estimation and compensation of the high-speed translational component from the radar echo are needed. The ultra-wideband signal model under the joint action of the target high-speed translational movement and micro-movement is proposed, and a state space approach based method for accurately estimating translational radial velocity is presented. This method can effectively inhibit the influence of the micro- movement component on the translational radial velocity estimation, and provide better estimation accuracy than traditional methods. The ultra-wideband signal model containing both high-speed translational components and micro-movement components can also be used directly for subsequent micro parameter extraction processing. The simulation results show the effectiveness of the signal model and the translational radial velocity estimation method presented in the paper.
In order to analyze the micro-movement component of the ballistic targets at the middle trajectory, the accurate estimation and compensation of the high-speed translational component from the radar echo are needed. The ultra-wideband signal model under the joint action of the target high-speed translational movement and micro-movement is proposed, and a state space approach based method for accurately estimating translational radial velocity is presented. This method can effectively inhibit the influence of the micro- movement component on the translational radial velocity estimation, and provide better estimation accuracy than traditional methods. The ultra-wideband signal model containing both high-speed translational components and micro-movement components can also be used directly for subsequent micro parameter extraction processing. The simulation results show the effectiveness of the signal model and the translational radial velocity estimation method presented in the paper.
2013, 35(2): 419-425.
doi: 10.3724/SP.J.1146.2012.00780
Abstract:
The electromagnetic scattering mechanism of radar target in high frequency domain can be characterized exactly by Geometrical Theory of Diffraction (GTD) model. In this paper, a novel parameter estimating method for 2D GTD model is proposed based on the analysis of the radar echoes sparse characteristic. The parameters estimation is converted to the issue of sparse signal reconstruction in the framework of Compressed Sensing (CS). In the proposed method, the signal support is first determined using 2D Fourier transform imaging and then the parameters of GTD model are estimated from the support region. To further improve the estimation precision of the parameters, clustering algorithms and linear least squares algorithms also adopted. Experiment results from both synthetic and real data show that the presented method is superior to the ones in existence, especially for the estimation of the scattering center type.
The electromagnetic scattering mechanism of radar target in high frequency domain can be characterized exactly by Geometrical Theory of Diffraction (GTD) model. In this paper, a novel parameter estimating method for 2D GTD model is proposed based on the analysis of the radar echoes sparse characteristic. The parameters estimation is converted to the issue of sparse signal reconstruction in the framework of Compressed Sensing (CS). In the proposed method, the signal support is first determined using 2D Fourier transform imaging and then the parameters of GTD model are estimated from the support region. To further improve the estimation precision of the parameters, clustering algorithms and linear least squares algorithms also adopted. Experiment results from both synthetic and real data show that the presented method is superior to the ones in existence, especially for the estimation of the scattering center type.
2013, 35(2): 426-431.
doi: 10.3724/SP.J.1146.2012.00879
Abstract:
The ship radiated noise is provided with a stable part of line-spectrum, and the underwater target can be effectively detected by using the line-spectrum signal. The technology of Passive Synthetic Aperture Sonar (PSAS), processing for space gain exchange by time gain, which is used for synthetic processing the receiving signal of small aperture array along the linear motion, thereby the detection ability of the weak target is more excellent. An improved PSAS algorithm is proposed in this paper, compared to conventional aperture synthesis algorithm, it does not require the array element to overlap, and it processes the received signals in a segment of tine by the overall valuation array element changes on the time domain and the spatial domain, then it can directly carry out delayed compensation. Through theoretical analysis and actual simulation on ocean-trial, the validity and feasibility of the algorithm are proved. Moreover, it is suitable enough for practical engineering.
The ship radiated noise is provided with a stable part of line-spectrum, and the underwater target can be effectively detected by using the line-spectrum signal. The technology of Passive Synthetic Aperture Sonar (PSAS), processing for space gain exchange by time gain, which is used for synthetic processing the receiving signal of small aperture array along the linear motion, thereby the detection ability of the weak target is more excellent. An improved PSAS algorithm is proposed in this paper, compared to conventional aperture synthesis algorithm, it does not require the array element to overlap, and it processes the received signals in a segment of tine by the overall valuation array element changes on the time domain and the spatial domain, then it can directly carry out delayed compensation. Through theoretical analysis and actual simulation on ocean-trial, the validity and feasibility of the algorithm are proved. Moreover, it is suitable enough for practical engineering.
2013, 35(2): 432-437.
doi: 10.3724/SP.J.1146.2012.00781
Abstract:
Effective channel order estimation is a critical step of blind channel identification. A blind channel effective order estimation algorithm based on SubSpace Channel Matrix Recursion (SS-CMR) is proposed. On the basis of the special structure of Q matrix and the its null space vector equivalently viewed as the convolution of true channel impulse response and the common zero channel, the channel matrix is obtained by SS method and then the estimated channel order is got by the constructed recursion cost function. The simulation proves that the performance of SS-CRM is improved compared with CMR and obviously better than the other existing algorithm, especially when the channel impulse response has small head and tail taps; analytical analysis shows that the algorithm complexity of SS-CMR is obviously reduced compared with CMR.
Effective channel order estimation is a critical step of blind channel identification. A blind channel effective order estimation algorithm based on SubSpace Channel Matrix Recursion (SS-CMR) is proposed. On the basis of the special structure of Q matrix and the its null space vector equivalently viewed as the convolution of true channel impulse response and the common zero channel, the channel matrix is obtained by SS method and then the estimated channel order is got by the constructed recursion cost function. The simulation proves that the performance of SS-CRM is improved compared with CMR and obviously better than the other existing algorithm, especially when the channel impulse response has small head and tail taps; analytical analysis shows that the algorithm complexity of SS-CMR is obviously reduced compared with CMR.
2013, 35(2): 438-444.
doi: 10.3724/SP.J.1146.2012.00517
Abstract:
A new adaptive digital beamforming in receiving end based on compressed sensing is proposed. In the case of sparse array antenna, receiving signal from absence elements can be reconstructed by using the theory of compressed sensing. Adaptive digital beamforming techniques are then adopted to form antenna beams, whose main lobe is steered to desired direction and nulls are steered to the directions of interferences. Simulation results with Monte Carlo method show that the beam performances of the proposed method are approaching to that of full array antenna, and actual antenna elements can be reduced greatly.
A new adaptive digital beamforming in receiving end based on compressed sensing is proposed. In the case of sparse array antenna, receiving signal from absence elements can be reconstructed by using the theory of compressed sensing. Adaptive digital beamforming techniques are then adopted to form antenna beams, whose main lobe is steered to desired direction and nulls are steered to the directions of interferences. Simulation results with Monte Carlo method show that the beam performances of the proposed method are approaching to that of full array antenna, and actual antenna elements can be reduced greatly.
2013, 35(2): 445-450.
doi: 10.3724/SP.J.1146.2012.00598
Abstract:
The lowest compressive sampling rate for non-uniform block sparse signals and the decay property of restricted isometry constant of measure matrix is theoretically analyzed. A blind recovery algorithm without knowing the order and distribution of blocks is proposed. The algorithm improves the estimation precision of nonzero values positions by dividing the block sparse signal uniformly for several times according to successive decreasing block length, and then eliminating the zero value positions in the uniform blocks using Orthogonal Matching Pursuit (OMP) method, which leads to a better recovery result. The performance of the blind recovery algorithm is analyzed and simulation results verify the effectiveness and practicality further.
The lowest compressive sampling rate for non-uniform block sparse signals and the decay property of restricted isometry constant of measure matrix is theoretically analyzed. A blind recovery algorithm without knowing the order and distribution of blocks is proposed. The algorithm improves the estimation precision of nonzero values positions by dividing the block sparse signal uniformly for several times according to successive decreasing block length, and then eliminating the zero value positions in the uniform blocks using Orthogonal Matching Pursuit (OMP) method, which leads to a better recovery result. The performance of the blind recovery algorithm is analyzed and simulation results verify the effectiveness and practicality further.
2013, 35(2): 451-456.
doi: 10.3724/SP.J.1146.2012.01087
Abstract:
This paper presents a novel integro-differential equation approach for removing Poisson noise. The classical Total Variational (TV) minimization model is discussed, and then the novel hierarchical multiscale variational image representation model is given. To arrive at the novel integro-differential equation, one integrates in inverse scale space a succession of refined slices of the image. The novel integro-differential equation includes a monotone increasing scaling function. According to choose an adaptive scaling function, this equation can remove Poisson noise efficiently. Finally, the experiment results demonstrate the proposed model obtains better effects compare with the classical TV and fourth-order partial differential equation models.
This paper presents a novel integro-differential equation approach for removing Poisson noise. The classical Total Variational (TV) minimization model is discussed, and then the novel hierarchical multiscale variational image representation model is given. To arrive at the novel integro-differential equation, one integrates in inverse scale space a succession of refined slices of the image. The novel integro-differential equation includes a monotone increasing scaling function. According to choose an adaptive scaling function, this equation can remove Poisson noise efficiently. Finally, the experiment results demonstrate the proposed model obtains better effects compare with the classical TV and fourth-order partial differential equation models.
A Transformation of Basic Probability Assignment to Probability Based on Conservation of Information
2013, 35(2): 457-462.
doi: 10.3724/SP.J.1146.2012.00208
Abstract:
The existing transform methods of basic probability assignment to probability are widespread lack?of?objective criteria and with much subjectivity. A linear weighted probability conversion method is presented. First, the?normalized priori information?is selected as weight to eliminate?the subjective factors?in transformation; Then,?an?equation?is constructed based on?the principle?of?information conservation before and after the transformation; Finally,?an iterative solution method is given for transformational probability. The experimental examples show that?the?solving speeds are fast,?the transformational probabilities are reasonable, effective, and?consistent with?the awareness level of the event.
The existing transform methods of basic probability assignment to probability are widespread lack?of?objective criteria and with much subjectivity. A linear weighted probability conversion method is presented. First, the?normalized priori information?is selected as weight to eliminate?the subjective factors?in transformation; Then,?an?equation?is constructed based on?the principle?of?information conservation before and after the transformation; Finally,?an iterative solution method is given for transformational probability. The experimental examples show that?the?solving speeds are fast,?the transformational probabilities are reasonable, effective, and?consistent with?the awareness level of the event.
2013, 35(2): 463-467.
doi: 10.3724/SP.J.1146.2012.00601
Abstract:
A novel supervised linear method based on Locality Preserving Projection (LPP) of reducing dimensionality is proposed for face recognition. In this study, the nearest neighbor graph of LPP is split into within-class graph and between-class graph according to the class label information of samples. After optimizing, the intrinsic local neighbor structure of the samples of same class is maintained and the distances between them are decreased. Meanwhile, the distances between the samples of different class are maximized to increase the space of the distribution of all kinds of samples, and thus the discriminability of the embedding is enhanced. In addition, adaptive neighborhood is applied to the construction of the graph, with the characterization for the sparsity of the sample improved. Experimental results on the two open face databases, Extended Yale B and CMU PIE face database, show that the proposed algorithm improves the accuracy of face recogntion effectively.
A novel supervised linear method based on Locality Preserving Projection (LPP) of reducing dimensionality is proposed for face recognition. In this study, the nearest neighbor graph of LPP is split into within-class graph and between-class graph according to the class label information of samples. After optimizing, the intrinsic local neighbor structure of the samples of same class is maintained and the distances between them are decreased. Meanwhile, the distances between the samples of different class are maximized to increase the space of the distribution of all kinds of samples, and thus the discriminability of the embedding is enhanced. In addition, adaptive neighborhood is applied to the construction of the graph, with the characterization for the sparsity of the sample improved. Experimental results on the two open face databases, Extended Yale B and CMU PIE face database, show that the proposed algorithm improves the accuracy of face recogntion effectively.
2013, 35(2): 468-473.
doi: 10.3724/SP.J.1146.2012.00791
Abstract:
Transformation electromagnetics, which has become a hotspot in the research areas recently, provides a new way to manipulate electromagnetic waves with coordinate transformation method. Combined with metamaterials, several new kinds of electromagnetic devices with peculiar functions can now be designed. In this paper, a series of polyline coordinate transformation devices are summarized, among which the electromagnetic wave concentrator is mainly discussed. Based on the theories, the scattering magnifying function and the military applications of electromagnetic wave concentrators are studied through the simulations of embedded stealth aircraft models and the calculations of their scattering cross sections. Each of the parameter extremums of this device and its normalized distributions are analyzed in detail at last.
Transformation electromagnetics, which has become a hotspot in the research areas recently, provides a new way to manipulate electromagnetic waves with coordinate transformation method. Combined with metamaterials, several new kinds of electromagnetic devices with peculiar functions can now be designed. In this paper, a series of polyline coordinate transformation devices are summarized, among which the electromagnetic wave concentrator is mainly discussed. Based on the theories, the scattering magnifying function and the military applications of electromagnetic wave concentrators are studied through the simulations of embedded stealth aircraft models and the calculations of their scattering cross sections. Each of the parameter extremums of this device and its normalized distributions are analyzed in detail at last.
2013, 35(2): 484-488.
doi: 10.3724/SP.J.1146.2012.00695
Abstract:
In order to assess multi distorted types of image quality effectively, a new no reference image quality assessment method is proposed, which uses General Regression Neural Network (GRNN) model to predict image quality score by learning phase congruency feature. In this method, three images, namely the Phase Congruency (PC) image, the maximum moment of PC covariance and the minimum moment of PC covariance image, are produced by the phase congruency model. Secondly the gradient entropy of the three reproduced images, the gradient mean value and the gradient entropy of the original image are computed by gray-level gradient co-occurrence matrix model, and the mean value of the three images are also calculated. At last all above eight features are fed to GRNN to learn, and predict image quality score. Experimental results demonstrate our algorithm is more consistent with human subjective scores and moreover has credible generalization.
In order to assess multi distorted types of image quality effectively, a new no reference image quality assessment method is proposed, which uses General Regression Neural Network (GRNN) model to predict image quality score by learning phase congruency feature. In this method, three images, namely the Phase Congruency (PC) image, the maximum moment of PC covariance and the minimum moment of PC covariance image, are produced by the phase congruency model. Secondly the gradient entropy of the three reproduced images, the gradient mean value and the gradient entropy of the original image are computed by gray-level gradient co-occurrence matrix model, and the mean value of the three images are also calculated. At last all above eight features are fed to GRNN to learn, and predict image quality score. Experimental results demonstrate our algorithm is more consistent with human subjective scores and moreover has credible generalization.
2013, 35(2): 489-493.
doi: 10.3724/SP.J.1146.2012.00812
Abstract:
A hyper-spectral image compression algorithm based on nonnegative tensor factorizations is proposed in this paper. First, every band of hyper-spectral images is decomposed by 2D 5/3 discrete wavelet transform to reduce the space redundancy of hyper-spectral images. Then, the four DWT sub-bands of the each level DWT for all spectral coverage are used as four tensors. And each sub-band tensor is decomposed by the proposed improved Hierarchical Alternating Least Squares (HALS) algorithm to reduce the spectra redundancy and the residual space redundancy. The algorithm can also protect the spectral information. Finally, the factorizations matrix is encoded by an entropy coder. The experimental results show that the proposed compression algorithm has good compressive property. In the compression ration range from 32:1 to 4:1, the average peak signal to noise ratio of proposed compression algorithm is higher than 40 dB. Compared with traditional approaches, the proposed method could improve the average PSNR by 1.499 dB. The compression performance of hyper-spectral image is effectively improved and the spectral information is protected.
A hyper-spectral image compression algorithm based on nonnegative tensor factorizations is proposed in this paper. First, every band of hyper-spectral images is decomposed by 2D 5/3 discrete wavelet transform to reduce the space redundancy of hyper-spectral images. Then, the four DWT sub-bands of the each level DWT for all spectral coverage are used as four tensors. And each sub-band tensor is decomposed by the proposed improved Hierarchical Alternating Least Squares (HALS) algorithm to reduce the spectra redundancy and the residual space redundancy. The algorithm can also protect the spectral information. Finally, the factorizations matrix is encoded by an entropy coder. The experimental results show that the proposed compression algorithm has good compressive property. In the compression ration range from 32:1 to 4:1, the average peak signal to noise ratio of proposed compression algorithm is higher than 40 dB. Compared with traditional approaches, the proposed method could improve the average PSNR by 1.499 dB. The compression performance of hyper-spectral image is effectively improved and the spectral information is protected.
2013, 35(2): 494-498.
doi: 10.3724/SP.J.1146.2012.00648
Abstract:
In order to improve the axial resolution and satisfy the contrast resolution of medical ultrasound imaging, a predistorted Linear Frequency Modulation (LFM) coded excitation new method based on amplitude weighting is proposed. Combining the amplitude weighting technology of LFM transmitted signal with the sidelobe reduction technology of echo signal, the method on the one hand can compensate the influence of transducer impulse response on transmission signal, so bandwidth of the echo signal is not limited by the transducer and axial resolution is improved. On the other hand, the method can remove the Fresnel ripples of transmitted signals frequency response, increase its bandwidth and use mismatched filter for pulse compression, so range sidelobe is suppressed to ensure contrast resolution. The results of simulation reveal that in contrast with constant envelope LFM coded excitation, the predistorted one can improve axial resolution and reduce maximum sidelobe at -48 dB to satisfy contrast resolution of medical imaging. FieldII Simulation results of B-mode image show that the axial resolution of constant envelope LFM coded excitation and the predistorted one is 0.35 mm and 0.25 mm, respectively.
In order to improve the axial resolution and satisfy the contrast resolution of medical ultrasound imaging, a predistorted Linear Frequency Modulation (LFM) coded excitation new method based on amplitude weighting is proposed. Combining the amplitude weighting technology of LFM transmitted signal with the sidelobe reduction technology of echo signal, the method on the one hand can compensate the influence of transducer impulse response on transmission signal, so bandwidth of the echo signal is not limited by the transducer and axial resolution is improved. On the other hand, the method can remove the Fresnel ripples of transmitted signals frequency response, increase its bandwidth and use mismatched filter for pulse compression, so range sidelobe is suppressed to ensure contrast resolution. The results of simulation reveal that in contrast with constant envelope LFM coded excitation, the predistorted one can improve axial resolution and reduce maximum sidelobe at -48 dB to satisfy contrast resolution of medical imaging. FieldII Simulation results of B-mode image show that the axial resolution of constant envelope LFM coded excitation and the predistorted one is 0.35 mm and 0.25 mm, respectively.
2013, 35(2): 499-503.
doi: 10.3724/SP.J.1146.2012.00869
Abstract:
A new Gray map is defined fromR=F2+uF2+u2F2+u3F2 toF42 with u4=0. It is proved that the Gray image of a linear (1+u+u2+u3)-cyclic code of lengthn over R is a distance-invariant linear cyclic code of length4n over F2. Further more, the generator polynomials of the Gray image of this constacyclic code for odd length over Ris determined, some optimal binary linear cyclic codes are also obtained.
A new Gray map is defined fromR=F2+uF2+u2F2+u3F2 toF42 with u4=0. It is proved that the Gray image of a linear (1+u+u2+u3)-cyclic code of lengthn over R is a distance-invariant linear cyclic code of length4n over F2. Further more, the generator polynomials of the Gray image of this constacyclic code for odd length over Ris determined, some optimal binary linear cyclic codes are also obtained.
2013, 35(2): 504-508.
doi: 10.3724/SP.J.1146.2012.00870
Abstract:
A resistanceR*out is in parallel with the output gap of the cavity by settingdiscrete portsin the three- dimensional microwave simulation software, the relationship between the resistance R*outand the gap-impedance real part in the klystron overall design requirements is given in this paper. Consequently, the fast and accurate design of the output circuit of filter is carried out only by just setting the optimization goals of the Voltage Standing Wave Ratio (VSWR) less than one constant, setting the size and position the inductance and capacitance for the optimization parameter.
A resistanceR*out is in parallel with the output gap of the cavity by settingdiscrete portsin the three- dimensional microwave simulation software, the relationship between the resistance R*outand the gap-impedance real part in the klystron overall design requirements is given in this paper. Consequently, the fast and accurate design of the output circuit of filter is carried out only by just setting the optimization goals of the Voltage Standing Wave Ratio (VSWR) less than one constant, setting the size and position the inductance and capacitance for the optimization parameter.
2013, 35(2): 474-483.
doi: 10.3724/SP.J.1146.2012.00931
Abstract:
Recently, a great number of interests have been attracted to perceptual video coding. In this paper, perceptual video coding and especially perceptual three-dimensional video coding are reviewed. Firstly, single-view video coding is analyzed by utilizing perceptual characteristics of human visual system including luminance, contrast sensitivity, fovea, etc, and then their problems are analyzed. Secondly, the perceptual three-dimensional video coding based on the region of interest and binocular sensitivity characteristics are discussed and analyzed. Finally, several future research directions for this technology and research field are discussed.
Recently, a great number of interests have been attracted to perceptual video coding. In this paper, perceptual video coding and especially perceptual three-dimensional video coding are reviewed. Firstly, single-view video coding is analyzed by utilizing perceptual characteristics of human visual system including luminance, contrast sensitivity, fovea, etc, and then their problems are analyzed. Secondly, the perceptual three-dimensional video coding based on the region of interest and binocular sensitivity characteristics are discussed and analyzed. Finally, several future research directions for this technology and research field are discussed.
