ID 原文 译文
24925 该模型以最小化网络总时延和能耗为目标，将拓扑设计问题转化为混合整数非线性规划模型的求解问题并求解出最优目标。 The model aims at minimizing the total network delay and energy consumption, and transforms the topology design problem into a mixed integer nonlinear programming model solution problem and solves the optimal goal．
24926 研究表明，网状拓扑结构具有较低网络时延、较高吞吐量以及较长的网络生存期，能满足肺部太赫兹纳米传感器网络的低能耗和低时延的要求。 The result shows that the mesh network topology has low network delay, high throughput and long network lifetime, which can meet the requirements of low energy consumption and low delay of lung terahertz nanosensor network．
24927 由于神经网络强大的学习能力与快速的运行速度，近年来基于深度学习的图像压缩感知(Image Com-pressive Sensing，ICS)研究备受关注。 Due to its great learning ability and fast processing speed, deep learning-based image compressive sensing (ICS) methods attract a lot of attention in recent years．
24928 然而，大多数现有 ICS 神经网络的结构设计忽略了传统迭代重构算法中的数学理论基础，无法有效利用信号中的先验结构知识，可解释性较差。 However, the design of most existing ICS neural networks architecture ignore the mathematical theory in iterative optimization-based methods and cannot effectively use the prior structure knowledge in the signal, leading to lack of the interpretability．
24929 为了保留优化算法核心思想并同时利用深度学习的高性能，本文使用可学习的卷积层替代了传统平滑投影 Landweber 算法(Smooth Projected Landweber，SPL)中的人工设计参数，提出一种新型 ICS 神经网络 SPLNet。 In order to retain the core ideas of the optimization algorithm and utilize the high performance of deep learning, this paper uses learnable convolutional layers to replace the predefined filters and artificial design parameters in the traditional smooth projected Landweber algorithm (SPL), and proposes a ICS neural network named SPLNet．
24930 在 SPLNet 中，设计了一个独特的网络结构 SPLBlock 实现 SPL 迭代过程中的三个核心步骤: (1)去除块效应的维纳滤波器; In SPLNet, we design a unique network structure SPLBlock to implement three key steps inSPL iteration: (1)Wiener filter for removal of blocking artifacts;
24931 (2)在凸投影集合上的近似操作; (2)approximation with projection onto the convex set;
24932 (3)实现稀疏表示及去噪的变换域双变量收缩。 (3)bivariate shrinkage on transform domain for sparse representation and denoising．
24933 仿真实验结果表明:与现有最优的 ICS 优化迭代算法 GSＲ 相比，SPLNet 的重构图像平均 PSNＲ 提升了0.78dB;与最优的神经网络框架 SCSNet 相比，SPLNet 的重构图像平均 PSNＲ 提升了 0. 92dB。 Experimental results indicate that, compared with current state-of-the-art ICS optimization iterative algorithm GSＲ, the average reconstructed image PSNＲ of SPL-Net are improved by 0. 78dB, and compared with state-of-the-art neural network framework SCSNet, the average reconstructed image PSNＲ of SPLNet are improved by 0.92dB．
24934 针对现有算法在卷积码参数识别过程中存在计算量大、容错性差的缺点，提出了一种基于迭代消元的快速识别方法。 In order to overcome the shortcomings of high computational complexity and low fault tolerance of the existing algorithms for recognition of convolutional codes, a fast convolutional code identification method based on iterative elimination is proposed．