为提高波达方向(Direction Of Arrival,DOA)的估计速度,该文基于子空间的正交性原理,利用噪声子空间及其共轭的交集进行奇异值分解(SVD)实现噪声子空间的降维,并基于降维噪声子空间与导向矢量及其共轭的双正交性提出一种2维阵列快速DOA...为提高波达方向(Direction Of Arrival,DOA)的估计速度,该文基于子空间的正交性原理,利用噪声子空间及其共轭的交集进行奇异值分解(SVD)实现噪声子空间的降维,并基于降维噪声子空间与导向矢量及其共轭的双正交性提出一种2维阵列快速DOA估计算法。理论分析和仿真实验表明:该算法不受实际阵型的限制,能将传统MUSIC谱的角度范围压缩至原来的一半,从而将DOA估计的计算量降至传统方法的50%,并具有与MUSIC算法相当的角度分辨率。展开更多
为利用互质结构进行二维高精度波达方向(direction of arrival,DOA)估计,设计了双平行互质阵列,提出了构建非均匀虚拟阵列的失配处理贝叶斯学习方法,最大限度扩展了测向自由度的同时,降低了网格失配对DOA估计精度的影响。首先,对平行互...为利用互质结构进行二维高精度波达方向(direction of arrival,DOA)估计,设计了双平行互质阵列,提出了构建非均匀虚拟阵列的失配处理贝叶斯学习方法,最大限度扩展了测向自由度的同时,降低了网格失配对DOA估计精度的影响。首先,对平行互质阵列进行垂直方向扩展构建了双平行互质阵列;其次,进行了非均匀虚拟阵列扩展,利用稀疏贝叶斯学习进行稀疏重构;然后,利用到达角相邻网格的能量关系,通过泰勒展开,进行了低复杂度的失配处理;最后,提出剔除规则和选择规则,融合两个方向子阵的估计结果。理论分析和仿真实验证明了所提阵列和DOA估计方法的有效性。展开更多
针对传统子空间算法需要进行特征值分解或奇异值分解等复杂计算的问题,提出一种双平行线阵(Double Parallel Linear Array,DPLA)的快速一维波达方向(Direction of Arrival,DOA)估计算法。算法通过处理互协方差矩阵的第一列元素构造出等...针对传统子空间算法需要进行特征值分解或奇异值分解等复杂计算的问题,提出一种双平行线阵(Double Parallel Linear Array,DPLA)的快速一维波达方向(Direction of Arrival,DOA)估计算法。算法通过处理互协方差矩阵的第一列元素构造出等效的噪声子空间,再通过求根MUSIC(Multiple Signal Classification)算法得到DOA估计,有效避开了特征值分解或奇异值分解,降低了计算复杂度,提高了运算速度。仿真结果表明,该算法在提高了估计精度的同时减少了估计时间。展开更多
Theoretical calculations of Double Hanging Ring Molecule(DHRM) [(GnHn-1^m)(GnHn-1^m)](G=C,Si,Ge;n=3,5,6,7,8;m=+1,-1,0,+1,+2) were performed via Gaussian 09 with the method of Density Functional Theory(DFT). Geometrica...Theoretical calculations of Double Hanging Ring Molecule(DHRM) [(GnHn-1^m)(GnHn-1^m)](G=C,Si,Ge;n=3,5,6,7,8;m=+1,-1,0,+1,+2) were performed via Gaussian 09 with the method of Density Functional Theory(DFT). Geometrical optimization, Potential Energy surface Scan(PES), Degree of Aromaticity(DOA) and Nucleus Independent Chemical Shift(NICS) were computed to study the optimal structures and aromaticity of DHRMs. Ring Stretching Vibration Raman Spectroscopy(RSVRSF) was predicted to seek the relation between RSVRSF and aromaticity of DHRMs. The results show optimal structures of DHRMs[(GnH(n-1)~m)(GnH(n–1)~m)](n = 3, 5~8);DA = 90° is the stable structure when n = 3, 7, 8;while n = 5 corresponds to DA = 30°, n = 6 corresponds to DA = 50°;the correlation between DOA and NICS of DHRMs is quadratic;the value of RSVRSF of DHRM approximates to its corresponding single ring molecule, which could act as characteristic frequency of ring molecule to identify its aromaticity;the correlation between RSVRSF and DOA is quadratic, and that between RSVRSF and NICS is linear.展开更多
文摘为提高波达方向(Direction Of Arrival,DOA)的估计速度,该文基于子空间的正交性原理,利用噪声子空间及其共轭的交集进行奇异值分解(SVD)实现噪声子空间的降维,并基于降维噪声子空间与导向矢量及其共轭的双正交性提出一种2维阵列快速DOA估计算法。理论分析和仿真实验表明:该算法不受实际阵型的限制,能将传统MUSIC谱的角度范围压缩至原来的一半,从而将DOA估计的计算量降至传统方法的50%,并具有与MUSIC算法相当的角度分辨率。
文摘为利用互质结构进行二维高精度波达方向(direction of arrival,DOA)估计,设计了双平行互质阵列,提出了构建非均匀虚拟阵列的失配处理贝叶斯学习方法,最大限度扩展了测向自由度的同时,降低了网格失配对DOA估计精度的影响。首先,对平行互质阵列进行垂直方向扩展构建了双平行互质阵列;其次,进行了非均匀虚拟阵列扩展,利用稀疏贝叶斯学习进行稀疏重构;然后,利用到达角相邻网格的能量关系,通过泰勒展开,进行了低复杂度的失配处理;最后,提出剔除规则和选择规则,融合两个方向子阵的估计结果。理论分析和仿真实验证明了所提阵列和DOA估计方法的有效性。
文摘针对传统子空间算法需要进行特征值分解或奇异值分解等复杂计算的问题,提出一种双平行线阵(Double Parallel Linear Array,DPLA)的快速一维波达方向(Direction of Arrival,DOA)估计算法。算法通过处理互协方差矩阵的第一列元素构造出等效的噪声子空间,再通过求根MUSIC(Multiple Signal Classification)算法得到DOA估计,有效避开了特征值分解或奇异值分解,降低了计算复杂度,提高了运算速度。仿真结果表明,该算法在提高了估计精度的同时减少了估计时间。
文摘利用集群自主式水下航行器(Autonomous Underwater Vehicles,AUV)进行的水下协同作业的需求越来越多。对于水下集群作业来说,AUV的水下定位非常重要。目前,AUV通常采用声学定位的工作模式,利用长、短基线阵对水下目标的二维波达方向(Direction of Arrival,DOA)进行估计,但在小型AUV上,基阵的阵列尺寸等受载体体积和换能器尺寸的共同限制,多信源条件下DOA估计的精度不高。设计低功耗平台,采用双平行线阵及传播算子算法来对多源目标进行二维DOA估计,结合通信与声学定位一体化方法,利用高频通信信号作为定位信号源,实现多信源环境中估计角度自动配对,在阵列尺寸受限的小型AUV上布放时,有较好的DOA估计效果。仿真结果验证了该方法的有效性。
基金Supported by the National Natural Science Foundation of China(No.21563023)the Graduate Education Innovation Program funded Projects of Inner Mongolia(No.S20161013506)inner Mongolia Normal University Graduate Students'Research&Innovation Fund(No.CXJJS16090)。
文摘Theoretical calculations of Double Hanging Ring Molecule(DHRM) [(GnHn-1^m)(GnHn-1^m)](G=C,Si,Ge;n=3,5,6,7,8;m=+1,-1,0,+1,+2) were performed via Gaussian 09 with the method of Density Functional Theory(DFT). Geometrical optimization, Potential Energy surface Scan(PES), Degree of Aromaticity(DOA) and Nucleus Independent Chemical Shift(NICS) were computed to study the optimal structures and aromaticity of DHRMs. Ring Stretching Vibration Raman Spectroscopy(RSVRSF) was predicted to seek the relation between RSVRSF and aromaticity of DHRMs. The results show optimal structures of DHRMs[(GnH(n-1)~m)(GnH(n–1)~m)](n = 3, 5~8);DA = 90° is the stable structure when n = 3, 7, 8;while n = 5 corresponds to DA = 30°, n = 6 corresponds to DA = 50°;the correlation between DOA and NICS of DHRMs is quadratic;the value of RSVRSF of DHRM approximates to its corresponding single ring molecule, which could act as characteristic frequency of ring molecule to identify its aromaticity;the correlation between RSVRSF and DOA is quadratic, and that between RSVRSF and NICS is linear.
