ID	原文	译文
1533	并同步分析了数据预处理、单个样本序列长度、网络结构参数等影响因素对分类准确率的影响，并针对典型探测场景分类进行了验证。	single sample sequencelength and network structure parameters on classification accuracy is analyzed synchronously, and verified for the typical de-tection scene classification.
1534	结果表明，LeNet 卷积神经网络在海上探测背景区分方面，具有很高的分类准确率，并且数据预处理方式、单个样本序列长度对结果影响显著，而网络结构参数有一定的调节区间，在此区间内调整，影响不显著，所提方法在顺/逆浪向、高/低海况条件下杂波分类与杂噪分类方面具有很高的准确率。	The application results of measured data show that the proposed method has high accuracy inclutter classification and noise classification under the conditions of forward/reverse direction and high/low sea conditions.
1535	本文首先讨论了大规模 MIMO-OFDM(Multiple-Input Multiple-Output Orthogonal Frequency Division Multi-plexing)系统信道的空间相关性，提出了一种基于隐聚类假设的信道建模方法，利用概率参数模拟不同的传播环境。	The paper discusses the spatial correlation of channels in massive MIMO-OFDM system, and proposes ahidden clustering hypothesis to simulations different propagation environments with probability parameters.
1536	然后，将机器学习领域的狄利特雷过程(Dirichlet Process，DP)引入到稀疏贝叶斯学习(Sparse Bayesian Learning，SBL)模型中，建立了 DP-SBL 结构，在信道估计的同时挖掘并利用大规模 MIMO 系统所特有的隐聚类特征。	Then, the Dirichlet process (DP)in machine learning is introduced into sparse Bayesian learning (SBL)model and a DP-SBL struc-ture is established. Consequently, the hidden clustering features of massive MIMO system are explored simultaneously in theprocess of channel estimation.
1537	接着，将 DP-SBL结构应用于大规模 MIMO-OFDM 系统中，在因子图上利用消息传递算法推导了一种基于隐聚类和狄利特雷过程的接收机算法。	Furthermore, the DP-SBL structure is applied to massive MIMO-OFDM systems, and a receiv-er algorithm based on hidden clustering and Dirichlet process is deduced by using message passing algorithm on factorgraphs.
1538	最后，将本文提出的接收机算法和现有算法进行对比分析。	Finally, we compare the proposed algorithm with the existing algorithms.
1539	结果表明，本文提出的接收机算法充分利用了大规模 MIMO-OFDM 系统特有的空间相关性，能够以较低的计算复杂度获得较强的鲁棒性和显著的性能增益。	Simulation results show that the proposed algorithm can exploit and utilize the spatial resources of massive MIMO-OFDM system. It can achieve remarkable perform-ance gain with low computation complexity and strong robustness.
1540	在基于压缩感知的阵列失效单元近场诊断方法中，使用结构化随机采样策略构造的观测矩阵约束等距特性未知，采用 1 范数极小化凸优化算法将无法确保阵列失效单元的高概率精确诊断。	The restricted isometry property of observation matrix in near-field measurements is unknown using ran-dom under-sampling strategy in compressed sensing based methods, which has a negative influence on the probability of suc-cess rate of diagnosis when adopting 1 norm minimization.
1541	针对这一不足，本文在深入研究非凸优化算法的基础上提出了一种基于随机扰动技术的非凸压缩感知近场诊断算法。	In order to overcome this limitation, a hybrid diagnosis algorithmusing random perturbation-non convex optimization for identification of impaired sensors in conformal arrays with near-fieldmeasurements is investigated.
1542	首先在失效单元个数满足稀疏性的前提下构造差异性阵列，	Differential array composed of healthy array and damaged array is constructed in the case of the sparsity of the number of failed elements.