ID 原文 译文
22655 仿真分析表明，该文算法对分布式信源进行直接位置估计的精度较传统直接位置估计算法明显提升，能够在较低信噪比下逼近克拉美罗界； The simulations show that the proposed methods have higher localization accuracy than traditional DPD methods when locating the distributed source, and are close to CRLB under the low SNR condition.
22656 分布源最大似然估计方法在低信噪比下定位精度优于广义子空间方法，而广义子空间方法复杂度更低。 DML-DPD method has higher localization accuracy than GS-DPD method in the case of low SNR, while GS-DPD method has less computational complexity than DML-DPD method.
22657 在大规模多输入多输出(MIMO)系统的上行链路检测算法中，最小均方误差(MMSE)算法是接近最优的，但算法涉及到大矩阵求逆运算，计算复杂度仍然较高。 Minimum Mean Square Error (MMSE) algorithm is near-optimal for uplink massive MIMO systems, but it involves high-complexity matrix inversion.
22658 近年提出的基于诺依曼级数近似的检测算法降低了复杂度但性能有一定的损失。 Recently, the proposed detection algorithm based on Neumann series approximation reduces the complexity with some performance losses.
22659 为了降低复杂度的同时逼近 MMSE 算法性能，该文提出基于二对角矩阵分解的诺依曼级数(Neumann Series)近似，即将大矩阵分解为以两条主对角线上元素组成的矩阵与空心矩阵之和。 In order to reduce the complexity while approaching the performance of MMSE algorithm, the Neumann series approximation based on two-diagonal matrix decomposition is proposed in this paper, that is, the large matrix is decomposed into the sum of the two elements of the main diagonal and the hollow matrix.
22660 理论分析与仿真结果表明所提算法检测性能逼近 MMSE 检测算法，且其复杂度从 ( )3O K 降低到 ( )2O K ，这里 K 是用户的数目。 The theoretical analysis and simulation results show that the detection performance of the proposed algorithm is close to the MMSE detection algorithm while its computational complexity is reduced from ( )3O K to ( )2O K , where K is the number of users.
22661 该文提出了一种基于 Hess 矩阵的多聚焦图像融合方法。 This paper proposes a Hess (also known as Hessian) matrix-based multi-focus image fusion method.
22662 该方法利用多尺度下的 Hess 矩阵检测特征和背景区域，并在此基础上，将源图像分成特征区域与非特征区域，分别采用不同的融合策略生成决策图； In this method, multi-scale Hess matrix is utilized to detect feature and background regions. On this basis, source images are split into two different parts, and different fusion strategies are applied to generating decision map respectively.
22663 然后通过结合不同部分的决策图，得到初始决策图；最后采用后处理方法对初始决策图进行精化，得到最终的融合图像。 By combining decision maps in different parts, an initial decision map is obtained, and then the initial decision map is refined with post-processing method.
22664 为了提高融合效果，该文还提出了一种基于多尺度 Hess 矩阵的聚焦评价方法。 To improve the performance of the fusion method, a new focus measure is proposed based on multi-scale Hess matrix for both feature and background regions.