ID 原文 译文
19635 针对无线虚拟化网络在时间域上业务请求的动态变化和信息反馈时延导致虚拟资源分配的不合理，该文提出一种基于长短时记忆(LSTM)网络的流量感知算法，该算法通过服务功能链(SFC)的历史队列信息来预测未来负载状态。 In order to solve the unreasonable virtual resource allocation caused by the dynamic change of service request and delay of information feedback in wireless virtualized network, a traffic-aware algorithm which exploits historical Service Function Chaining (SFC) queue information to predict future load state based onLong Short-Term Memory (LSTM) network is proposed.
19636 基于预测的结果，联合考虑虚拟网络功能(VNF)的调度问题和相应的计算资源分配问题，提出一种基于最大最小蚁群算法(MMACA)的虚拟网络功能动态部署方法，在满足未来队列不溢出的最低资源需求的前提下，采用按需分配的方式最大化计算资源利用率。 With the prediction results, the Virtual NetworkFunction (VNF) deployment and the corresponding computing resource allocation problems are studied, and aVNFs’ deployment method based on Maximum and Minimum Ant Colony Algorithm (MMACA) is developed.On the premise of satisfying the minimum resource demand for future queue non-overflow, the on-demandallocation method is used to maximize the computing resource utilization.
19637 仿真结果表明，该文提出的基于LSTM神经网络预测模型能够获得很好的预测效果，实现了网络的在线监测； Simulation results show that theprediction model based on LSTM neural network in this paper obtains good prediction results and realizesonline monitoring of the network.
19638 基于MMACA的VNF部署方法有效降低了比特丢失率的同时也降低了整体VNF调度产生的平均端到端时延。 The Maximum and Minimum Ant Colony Algorithm based VNF deploymentmethod reduces effectively the bit loss rate and the average end-to-end delay caused by overall VNFs’scheduling at the same time.
19639 针对现有频域显著性检测方法得到的显著区域不完整的问题，该文提出一种多尺度分析的频率域显著性检测方法。 To solve the incompleteness of the salient region obtained by the existing saliency detection methodin the frequency domain, a frequency saliency detection method of multi-scale analysis is proposed.
19640 首先由输入图像特征通道信息构建4元超复数，然后通过小波变换对4元超复数域中幅度谱进行多尺度分解，计算生成多尺度下的视觉显著图，最后由评价函数选出效果较好显著图合成最终视觉显著图。 Firstly, thequaternion hypercomplex is constructed by the input image feature channels. Then, the multi-scaledecomposition of the quaternion amplitude spectrum is performed by wavelet transform, and the multi-scalevisual saliency map is calculated. Finally, the better saliency map is fused based on the evaluation function, andcentral bias is used to generate the final visual saliency map.
19641 实验结果表明，该文方法能够有效地抑制背景干扰，快速、精确地找到完整的显著目标，具有较高的检测精确度。 The experimental results show that the proposed method can effectively suppress the background interference, find significant target quickly and accurately, and have high detection accuracy.
19642 针对目前高斯消元法在归零Turbo码长、帧同步等参数识别过程存在容错性能低且计算复杂度高的缺点，该文提出一种低信噪比(SNR)下基于差分似然差(DLD)的识别算法。 In order to overcome the shortcomings of low fault-tolerance and high computational complexity inthe process of parameter identification such as code length and synchronization of Turbo code, a new algorithmbased on Differential Likelihood Difference (DLD) at low Signal-to-Noise Ratio (SNR) is proposed.
19643 首先通过定义差分似然差的概念，利用归零Turbo码帧头两码元差分似然差为正值(“+”)的特性，构建分析矩阵实现码长的识别； Firstly, theconcept of DLD is defined, and the analysis matrix is constructed to identify the code length by using thecharacteristic that the DLD between two codes in Turbo frame terminal is positive ("+");
19644 其次，提出基于最小错误判决准则下的差分似然差“+”位置门限判决方法，完成帧同步； Secondly, a methodbased on the minimum error decision criterion to decide DLD "+" position is proposed to complete framesynchronization.