This paper explores the recovery of block sparse signals in frame-based settings using the l_(2)/l_(q)-synthesis technique(0<q≤1).We propose a new null space property,referred to as block D-NSP_(q),which is based ...This paper explores the recovery of block sparse signals in frame-based settings using the l_(2)/l_(q)-synthesis technique(0<q≤1).We propose a new null space property,referred to as block D-NSP_(q),which is based on the dictionary D.We establish that matrices adhering to the block D-NSP_(q)condition are both necessary and sufficient for the exact recovery of block sparse signals via l_(2)/l_(q)-synthesis.Additionally,this condition is essential for the stable recovery of signals that are block-compressible with respect to D.This D-NSP_(q)property is identified as the first complete condition for successful signal recovery using l_(2)/l_(q)-synthesis.Furthermore,we assess the theoretical efficacy of the l2/lq-synthesis method under conditions of measurement noise.展开更多
This paper presents a simple method of forming sum and difference patterns with adaptivenulls. The effects on the sidelobe level and the pointing null of difference pattern by adaptive null areanalyzed. The result tha...This paper presents a simple method of forming sum and difference patterns with adaptivenulls. The effects on the sidelobe level and the pointing null of difference pattern by adaptive null areanalyzed. The result that the increment value of the envelope of the sidelobe level under the effect ofa null is less than l.6dB is proved. The formula about shift value of the pointing null which is thefunction of the jammer direction and array parameters is given in the paper.展开更多
The high-pressure electro-pneumatic servo valve(HESV)is a core element of the high-pressure pneumatic servo system.The annular clearance and the rounded corner of the spool-sleeve can cause the leakage at null positio...The high-pressure electro-pneumatic servo valve(HESV)is a core element of the high-pressure pneumatic servo system.The annular clearance and the rounded corner of the spool-sleeve can cause the leakage at null position,thereby affecting high-precision control and stability of the servo system.This paper investigates the effects of the clearance structure on leakage behavior at null position of the HESV.A numerical approach was employed to evaluate the effects,and then a mathematical model was established to obtain the variation law of leakage flow rate at null position.The results indicate that the leakage flow rate at null position varies linearly with supply pressure and rounded corner radius,and is nonlinear as a quadratic concave function with annular clearance.The leakage flow rate of the annular clearance and the rounded corner varies with the valve opening in an invariable−nonlinear−linear trend.A test rig system of leakage behavior at null position of the HESV was built to confirm the validity of the numerical model,which agrees well with the conducted experimental study.展开更多
Dwarfing is useful to reduce plant height,when breeding high-yielding and non-lodging crops.In this study,a set of natural storage protein subunit-null dwarf mutants of soybean was reported that showed strongly reduce...Dwarfing is useful to reduce plant height,when breeding high-yielding and non-lodging crops.In this study,a set of natural storage protein subunit-null dwarf mutants of soybean was reported that showed strongly reduced plant stature and deficiency in various 7S and 11S subunits,designated as snd1 mutants.Under normal growth conditions,the snd1 mutants showed a severe dwarf phenotype,with plant height of about 25 cm.Compared with wild-type DN47,the mutant snd1 exhibited no obvious morphological differences at the early stage of development.All the snd1 mutants examined had fewer nodes and shorter than normal internodes;the leaves were similar in shape to normal parents,but were dark-green at the mature stage.The flower size was similar to DN47;however,the flowering period was shorter than in the wild-type.Significant variation was noted for protein content,oil content of the seeds and size of seeds(weight of 100 seeds)among 17 snd1 dwarf lines.Genetic analysis indicated that the dwarfism of snd1 was controlled by a single recessive gene.The snd1 dwarf mutant had markedly different dynamic levels of the endogenous hormones gibberellin(GA),brassinosteroid,indole-3-acetic acid and abscisic acid,at the seedling stage.Exogenous GA3 treatment led to recovery of the plant height phenotype of the snd1 mutant;GA3 at 0.1 mm had the largest effect on enhancing plant height.Using molecular markers,snd1 gene was approximately mapped in an interval of 603 kb between markers Satt166 and Satt561 on chromosome 19.Snd1 mutant provided valuable material for hypoallergenic soybean breeding and the snd1 gene might be a novel gene related to plant height in soybean.展开更多
雷达、声呐和无线通信等应用对于自适应波束形成的抗干扰能力和实时性提出了更高的要求。传统基于最速迭代的自适应波束形成算法存在“过拟合”特性,导致在相干干扰条件下的干扰抑制性能急剧下降。另外,当干扰存在扰动且导向向量失配时...雷达、声呐和无线通信等应用对于自适应波束形成的抗干扰能力和实时性提出了更高的要求。传统基于最速迭代的自适应波束形成算法存在“过拟合”特性,导致在相干干扰条件下的干扰抑制性能急剧下降。另外,当干扰存在扰动且导向向量失配时,也无法有效抑制干扰。针对上述问题,本文提出了一种基于共轭梯度(Conjugate Gradient,CG)加速的二次约束宽零陷干扰抑制自适应波束形成方法。该方法首先利用CG算法的快速收敛特性,完成采样协方差矩阵与导向向量间线性方程组的求解;其次将CG算法输出的权矢量作为迭代最速波束形成方法的初始权值,利用该方法的“过拟合”特性,确保对期望信号的强锁定;最后提出了一种强化干扰特征的波达方向(Direction of Arrival,DOA)估计方法,实现宽带相干干扰下的干扰来波方向估计,并将该方法与二次约束零陷展宽方法结合,用于捕获干扰特征,形成自适应零陷。仿真实验验证了所提方法在单快拍、宽带相干干扰条件下,能够自适应抑制干扰且稳健性较好。展开更多
图像压缩感知能从低采样观测中重建出高质量图像。将深度学习应用于图像压缩感知,可显著提高图像重建质量。然而,基于深度学习的图像压缩感知方法存在模型可解释性差、结构盲目设计而影响重建性能的问题。针对这些问题,提出了一种基于...图像压缩感知能从低采样观测中重建出高质量图像。将深度学习应用于图像压缩感知,可显著提高图像重建质量。然而,基于深度学习的图像压缩感知方法存在模型可解释性差、结构盲目设计而影响重建性能的问题。针对这些问题,提出了一种基于零值域分解的深度图像压缩感知方法(range-null space decomposition based deep image compressive sensing network,RND-Net)。该方法通过全局卷积采样的方式稀疏感知图像的特征信息,通过学习信号相关的采样矩阵,使采样值包含更丰富的图像特征,且相较一般的逐块采样方式,在全局层面上的采样可明显减少块状伪影;基于零值域分解的数学表示,将采样与重建过程转化为端到端深度学习模型,借助深度神经网络拟合所涉及的线性或非线性运算,相比传统方法缩短了模型推理时间,提升了图像重建能力。上述将数学先验知识有效融入数据驱动的方法称为协同驱动,既充分利用了数学先验知识,强化了模型的可解释性,使模型结构更易于设计,又发挥了以深度学习为代表的数据驱动方法的自主寻优能力,相比其他深度压缩感知方法更易于获得全局最优解。在多个测试集上的实验证明,RND-Net与目前图像重建能力较好的算法相比显著提升了图像重建质量,减少了单幅图像重建时间。当采样率为0.1、测试集为BSDS68时,RND-Net比AutoBCS在峰值信噪比(PSNR)上平均高1.02 dB。在测试集Set14上,RND-Net对于混合驱动的GPX-ADMM-Net的平均PSNR和结构相似性指数(SSIM)增益分别为1.15dB和0.0518;重建单幅图像时,RND-Net比GPX-ADMM-Net快约0.1049 s。展开更多
基金Supported by The Featured Innovation Projects of the General University of Guangdong Province(2023KTSCX096)The Special Projects in Key Areas of Guangdong Province(ZDZX1088)Research Team Project of Guangdong University of Education(2024KYCXTD018)。
文摘This paper explores the recovery of block sparse signals in frame-based settings using the l_(2)/l_(q)-synthesis technique(0<q≤1).We propose a new null space property,referred to as block D-NSP_(q),which is based on the dictionary D.We establish that matrices adhering to the block D-NSP_(q)condition are both necessary and sufficient for the exact recovery of block sparse signals via l_(2)/l_(q)-synthesis.Additionally,this condition is essential for the stable recovery of signals that are block-compressible with respect to D.This D-NSP_(q)property is identified as the first complete condition for successful signal recovery using l_(2)/l_(q)-synthesis.Furthermore,we assess the theoretical efficacy of the l2/lq-synthesis method under conditions of measurement noise.
文摘This paper presents a simple method of forming sum and difference patterns with adaptivenulls. The effects on the sidelobe level and the pointing null of difference pattern by adaptive null areanalyzed. The result that the increment value of the envelope of the sidelobe level under the effect ofa null is less than l.6dB is proved. The formula about shift value of the pointing null which is thefunction of the jammer direction and array parameters is given in the paper.
基金Project(51705164)supported by the National Natural Science Foundation of China。
文摘The high-pressure electro-pneumatic servo valve(HESV)is a core element of the high-pressure pneumatic servo system.The annular clearance and the rounded corner of the spool-sleeve can cause the leakage at null position,thereby affecting high-precision control and stability of the servo system.This paper investigates the effects of the clearance structure on leakage behavior at null position of the HESV.A numerical approach was employed to evaluate the effects,and then a mathematical model was established to obtain the variation law of leakage flow rate at null position.The results indicate that the leakage flow rate at null position varies linearly with supply pressure and rounded corner radius,and is nonlinear as a quadratic concave function with annular clearance.The leakage flow rate of the annular clearance and the rounded corner varies with the valve opening in an invariable−nonlinear−linear trend.A test rig system of leakage behavior at null position of the HESV was built to confirm the validity of the numerical model,which agrees well with the conducted experimental study.
基金Supported by the Ministry of Science and Technology of China(2016YFD0100500)Funding from Harbin Science and Technology Bureau(2016RQYXJ018,2017RAQXJ104)+4 种基金the Key Laboratory of Soybean Biology in the Chinese Ministry of Education,Northeast Agricultural University(SB17A01)National Natural Science Foundation of China(31801386)Heilongjiang Natural Science Foundation(LC2018008)Heilongjiang General Young Innovative Talents Training Plan(UNPYSCT-2018158)Certificate of China Postdoctoral Science Foundation Grant(2018M641839)
文摘Dwarfing is useful to reduce plant height,when breeding high-yielding and non-lodging crops.In this study,a set of natural storage protein subunit-null dwarf mutants of soybean was reported that showed strongly reduced plant stature and deficiency in various 7S and 11S subunits,designated as snd1 mutants.Under normal growth conditions,the snd1 mutants showed a severe dwarf phenotype,with plant height of about 25 cm.Compared with wild-type DN47,the mutant snd1 exhibited no obvious morphological differences at the early stage of development.All the snd1 mutants examined had fewer nodes and shorter than normal internodes;the leaves were similar in shape to normal parents,but were dark-green at the mature stage.The flower size was similar to DN47;however,the flowering period was shorter than in the wild-type.Significant variation was noted for protein content,oil content of the seeds and size of seeds(weight of 100 seeds)among 17 snd1 dwarf lines.Genetic analysis indicated that the dwarfism of snd1 was controlled by a single recessive gene.The snd1 dwarf mutant had markedly different dynamic levels of the endogenous hormones gibberellin(GA),brassinosteroid,indole-3-acetic acid and abscisic acid,at the seedling stage.Exogenous GA3 treatment led to recovery of the plant height phenotype of the snd1 mutant;GA3 at 0.1 mm had the largest effect on enhancing plant height.Using molecular markers,snd1 gene was approximately mapped in an interval of 603 kb between markers Satt166 and Satt561 on chromosome 19.Snd1 mutant provided valuable material for hypoallergenic soybean breeding and the snd1 gene might be a novel gene related to plant height in soybean.
文摘雷达、声呐和无线通信等应用对于自适应波束形成的抗干扰能力和实时性提出了更高的要求。传统基于最速迭代的自适应波束形成算法存在“过拟合”特性,导致在相干干扰条件下的干扰抑制性能急剧下降。另外,当干扰存在扰动且导向向量失配时,也无法有效抑制干扰。针对上述问题,本文提出了一种基于共轭梯度(Conjugate Gradient,CG)加速的二次约束宽零陷干扰抑制自适应波束形成方法。该方法首先利用CG算法的快速收敛特性,完成采样协方差矩阵与导向向量间线性方程组的求解;其次将CG算法输出的权矢量作为迭代最速波束形成方法的初始权值,利用该方法的“过拟合”特性,确保对期望信号的强锁定;最后提出了一种强化干扰特征的波达方向(Direction of Arrival,DOA)估计方法,实现宽带相干干扰下的干扰来波方向估计,并将该方法与二次约束零陷展宽方法结合,用于捕获干扰特征,形成自适应零陷。仿真实验验证了所提方法在单快拍、宽带相干干扰条件下,能够自适应抑制干扰且稳健性较好。
文摘图像压缩感知能从低采样观测中重建出高质量图像。将深度学习应用于图像压缩感知,可显著提高图像重建质量。然而,基于深度学习的图像压缩感知方法存在模型可解释性差、结构盲目设计而影响重建性能的问题。针对这些问题,提出了一种基于零值域分解的深度图像压缩感知方法(range-null space decomposition based deep image compressive sensing network,RND-Net)。该方法通过全局卷积采样的方式稀疏感知图像的特征信息,通过学习信号相关的采样矩阵,使采样值包含更丰富的图像特征,且相较一般的逐块采样方式,在全局层面上的采样可明显减少块状伪影;基于零值域分解的数学表示,将采样与重建过程转化为端到端深度学习模型,借助深度神经网络拟合所涉及的线性或非线性运算,相比传统方法缩短了模型推理时间,提升了图像重建能力。上述将数学先验知识有效融入数据驱动的方法称为协同驱动,既充分利用了数学先验知识,强化了模型的可解释性,使模型结构更易于设计,又发挥了以深度学习为代表的数据驱动方法的自主寻优能力,相比其他深度压缩感知方法更易于获得全局最优解。在多个测试集上的实验证明,RND-Net与目前图像重建能力较好的算法相比显著提升了图像重建质量,减少了单幅图像重建时间。当采样率为0.1、测试集为BSDS68时,RND-Net比AutoBCS在峰值信噪比(PSNR)上平均高1.02 dB。在测试集Set14上,RND-Net对于混合驱动的GPX-ADMM-Net的平均PSNR和结构相似性指数(SSIM)增益分别为1.15dB和0.0518;重建单幅图像时,RND-Net比GPX-ADMM-Net快约0.1049 s。
