在大规模多输入多输出系统中,最小均方误差(minimum mean square error,MMSE)算法能达到接近最优的线性信号检测性能,但是MMSE算法需要复杂的矩阵求逆运算,这限制了该算法的应用。为了降低运算复杂度,改进MMSE算法,利用Barzilai-Borwein...在大规模多输入多输出系统中,最小均方误差(minimum mean square error,MMSE)算法能达到接近最优的线性信号检测性能,但是MMSE算法需要复杂的矩阵求逆运算,这限制了该算法的应用。为了降低运算复杂度,改进MMSE算法,利用Barzilai-Borwein(BB)迭代算法来避免矩阵求逆运算,提出了结构简单的BB迭代信号检测算法,且基于信道硬化特性进一步优化了迭代初始解以加快算法的收敛速度。理论和仿真结果表明,所提出的BB迭代算法的性能优于最近提出的Neumann级数展开算法,而其复杂度相比截短阶数i=3的Neumann级数展开算法减少了一个数量级;且该算法收敛速度较快,在给定初始值的条件下,通过简单的几次迭代,能够快速接近MMSE算法的检测性能。展开更多
针对数字锁定放大器中稀疏分解算法检测精度受限的缺点,结合非线性方法进行补偿,采用混沌检测联合稀疏分解的算法设计了数字锁定放大器。数字锁定放大器借助稀疏分解算法估计信号参数,根据估计得到的参数设计对应混沌系统;利用混沌系统...针对数字锁定放大器中稀疏分解算法检测精度受限的缺点,结合非线性方法进行补偿,采用混沌检测联合稀疏分解的算法设计了数字锁定放大器。数字锁定放大器借助稀疏分解算法估计信号参数,根据估计得到的参数设计对应混沌系统;利用混沌系统检测信号的过程中产生间歇性混沌状态,并通过测量间歇性混沌状态的周期,精确检测信号。最终,在信号的信噪比低至-20 d B时可以成功检测信号,频率精度达到0.05 Hz,提供了一种非线性微弱信号检测方法。展开更多
A newly developed Deep Ocean Compact Autonomous Raman Spectrometer （DOCARS） system is introduced and used for in-situ detection of acid radical ions in this paper. To evaluate the feasibility and capability of DOCAR...A newly developed Deep Ocean Compact Autonomous Raman Spectrometer （DOCARS） system is introduced and used for in-situ detection of acid radical ions in this paper. To evaluate the feasibility and capability of DOCARS for quantitative analysis of the acid radical ions in the deep ocean, extensive investigations have been carried out both in laboratory and sea trials during the development phase. In the laboratory investigations, Raman spectra of the prepared samples （acid radical ions solutions） were obtained, and analyzed using the method of internal standard normalization in data processing. The Raman signal of acid radical ions was normalized by that of water molecules. The calibration curve showed that the normalized Raman signal intensity of SO4^2-, NO3^-, and HCO^-3 increases linearly as the concentration rises with correlation coefficient R^2 of 0.99, 0.99, and 0.98 respectively. The linear function obtained from the calibration curve was then used for the analysis of the spectra ,data acquired in the sea trial under a simulating chemical field in the deep-sea environment. It was found that the detected concentration of NO3 according to the linear function can reflect the concentration changes of NO~ after the sample was released, and the detection accuracy of the DOCARS system for SO^2-_4 is 8%. All the results showed that the DOCARS system has great potential in quantitative detection of acid radical ions under the deep-sea environment, while the sensitivity of the DOCARS system is expected to be improved.展开更多
文摘在大规模多输入多输出系统中,最小均方误差(minimum mean square error,MMSE)算法能达到接近最优的线性信号检测性能,但是MMSE算法需要复杂的矩阵求逆运算,这限制了该算法的应用。为了降低运算复杂度,改进MMSE算法,利用Barzilai-Borwein(BB)迭代算法来避免矩阵求逆运算,提出了结构简单的BB迭代信号检测算法,且基于信道硬化特性进一步优化了迭代初始解以加快算法的收敛速度。理论和仿真结果表明,所提出的BB迭代算法的性能优于最近提出的Neumann级数展开算法,而其复杂度相比截短阶数i=3的Neumann级数展开算法减少了一个数量级;且该算法收敛速度较快,在给定初始值的条件下,通过简单的几次迭代,能够快速接近MMSE算法的检测性能。
文摘针对数字锁定放大器中稀疏分解算法检测精度受限的缺点,结合非线性方法进行补偿,采用混沌检测联合稀疏分解的算法设计了数字锁定放大器。数字锁定放大器借助稀疏分解算法估计信号参数,根据估计得到的参数设计对应混沌系统;利用混沌系统检测信号的过程中产生间歇性混沌状态,并通过测量间歇性混沌状态的周期,精确检测信号。最终,在信号的信噪比低至-20 d B时可以成功检测信号,频率精度达到0.05 Hz,提供了一种非线性微弱信号检测方法。
基金Supported by the National High Technology Research and Development Program of China(863 Program)(Nos.2006AA09Z243,2012AA09A405)
文摘A newly developed Deep Ocean Compact Autonomous Raman Spectrometer （DOCARS） system is introduced and used for in-situ detection of acid radical ions in this paper. To evaluate the feasibility and capability of DOCARS for quantitative analysis of the acid radical ions in the deep ocean, extensive investigations have been carried out both in laboratory and sea trials during the development phase. In the laboratory investigations, Raman spectra of the prepared samples （acid radical ions solutions） were obtained, and analyzed using the method of internal standard normalization in data processing. The Raman signal of acid radical ions was normalized by that of water molecules. The calibration curve showed that the normalized Raman signal intensity of SO4^2-, NO3^-, and HCO^-3 increases linearly as the concentration rises with correlation coefficient R^2 of 0.99, 0.99, and 0.98 respectively. The linear function obtained from the calibration curve was then used for the analysis of the spectra ,data acquired in the sea trial under a simulating chemical field in the deep-sea environment. It was found that the detected concentration of NO3 according to the linear function can reflect the concentration changes of NO~ after the sample was released, and the detection accuracy of the DOCARS system for SO^2-_4 is 8%. All the results showed that the DOCARS system has great potential in quantitative detection of acid radical ions under the deep-sea environment, while the sensitivity of the DOCARS system is expected to be improved.
