ID 原文 译文
993 为了进一步提高 ADMM 惩罚译码速度，通过利用 Wei 等人提出的方法来减少欧几里德投影的次数，本文设计了基于 I-l1-PF 罚函数的水平分层调度与垂直分层调度策略的两种 LDPC 码 ADMM 惩罚译码方法。 In order to increase the ADMM penalized decoding speedfurther, by using the method proposed by Wei et al to reduce the number of Euclidean projections, this paper designs twokinds of ADMM penalized decoding methods with the horizontal layered scheduling and the vertical layered scheduling strat-egy for LDPC codes based on the I-l1-PF penalty function.
994 仿真实验表明，与现有 ADMM 惩罚译码方法相比较，所设计的译码方法不仅具有较好的译码性能，而且能够显著降低 LDPC 码译码的平均迭代次数和平均译码时间。 Simulation results show that the designed methods not only havebetter decoding performance but also significantly reduce the average number of iterations and the average decoding timecompared with the existing ADMM penalized decoding methods.
995 针对现有的辐射源识别方法的不足，利用辐射源自身 I/Q 失配指纹特征的可检测、可重现、不变的特性，提出一种基于 I/Q 失配的新型射频指纹特征提取方法。 In view of the shortcomings of the existing method of identification of radiant sources, a novel feature ex-traction method based on I/Q mismatch is proposed, which can be detected, reproducible and invariable by using the I/Qmismatch fingerprint of the source.
996 该方法是一种基于信号空间的指纹特征提取方法， This method is a fingerprint feature extraction method based on signal space.
997 它需要对信号的信噪比进行估计。 It needs to estimate the signal-to-noise ratio of the signal.
998 综合考虑，采用基于特征值分解的信噪比估计方法。 Through comprehensive consideration, the SNR estimation method based oneigenvalue decomposition is adopted.
999 实验仿真结果表明，在相同的信噪比条件下，基于 I/Q 失配的射频指纹特征提取方法比双谱、Hilbert-Huang 变换信号特征提取方法具有更加优越的射频指纹识别性能。 The experimental simulation results show that, under the same signal to noise ratio, theradio frequency fingerprint feature extraction method based on I/Q mismatch is superior to the dual spectrum and Hilbert-Huang transform signal feature extraction methods.
1000 针对非限定性条件下人脸识别困难问题，设计了一种引入感知模型的改进孪生卷积神经网络结构(In-ception Module Incorporated Siamese Convolutional Neural Networks，IMISCNN)， To solve the problem of face recognition under unlimited conditions, a simple network structure named In-ception Module Incorporated Siamese Convolutional Neural Networks (IMISCNN)was designed, which was suitable forsmall-scale data sets.
1001 在充分利用孪生结构有效减少外界干扰并避免过拟合等优点的基础上，为其增加感知模型实现更丰富特征的提取。 On the basis of making full use of the Siamese structure to effectively reduce external interference andavoid over-fitting, inception module was incorporated to the Siamese network to extract richer features.
1002 为寻找最优学习率引入了循环学习率策略，加速模型收敛。 Furthermore, a cycli-cal learning rate strategy was adopted to accelerate the convergence of the model.