基于SWIPT的Polar编码协作码的设计与系统性能研究

极化（Polar）编码协作可同时获得编码增益与分集增益,实现可靠通信。为了解决中继能量受限问题,本文研究了基于无线信息与能量同传（Simultaneous Wireless Information and Power Transfer, SWIPT）技术的Polar编码协作系统。首先,建立了基于SWIPT的Polar编码协作系统模型。其次,鉴于Polar码的Plotkin构造方法将一个长码分裂成两个短码,非常适用于编码协作场景,使用Plotkin构造方法联合设计信源节点和中继节点Polar码,并在目的节点对其进行联合串行相消（Successive Cancellation, SC）译码。相比传统点对点系统,基于SWIPT的Polar编码协作系统使得系统中断概率大幅度降低;与随机低密度奇偶校验（Low Density Parity Check, LDPC）编码协作系统相比,在译码迭代次数低时所提方案误码性能更佳。      

基于DeepLabV3与GAN的雷达时频混叠多信号智能检测与分离

针对当前雷达信号智能检测研究中存在信号数量受限、样式单一、不便于后续处理等问题,本文提出了一种基于DeepLabV3与GAN的雷达时频混叠多信号智能检测与分离方法。首先对接收信号的时域数据通过时频变换得到二维的时频数据,利用DeepLabV3对多信号重叠的时频数据进行检测并且实现信号的分离,对于信号的重叠部分利用GAN对时域信号进行估计重构。实验结果表明,该方法在信噪比（SNR）为-3 dB时,检测平均交并比（mIoU）能够达到了85%以上,在SNR为6 dB时,分离后的信号与原信号相关系数高于0.85。      

静电喷印技术在太赫兹超表面制备中的应用

近年来,太赫兹技术快速发展,基于超表面的太赫兹器件受到广泛关注,并已应用于太赫兹成像、光谱和生物传感等诸多领域。但太赫兹超表面器件的制备复杂且成本高,而静电喷印技术无需掩模版,成本低、精度高且易于在异形曲面上制作。基于静电喷印技术设计和制备了太赫兹吸波器,并利用太赫兹时域光谱系统（THz-TDS）对样品进行了性能测试,实验与仿真结果基本相符,在0.098~0.353 THz频段内吸收大于90%。此外,还设计了太赫兹线极化转换器,在0.167~0.355 THz频段内的极化转换大于95%,相对带宽约72%,并分析了所设计样品的制备工艺条件,验证了静电喷印技术对于制备太赫兹极化转换器的可行性。研究成果表明,静电喷印技术在太赫兹超表面器件的制备中具有广泛的应用前景。     

高效学习的透过未知散射介质的相位恢复方法

透过散射介质对目标进行准确的重建仍然是阻碍人们对深层生物组织成像分析和深空天文观测的主要挑战之一。基于深度学习的散射计算成像方法虽然在成像质量和效率等方面取得了很大的进展,但是针对实际系统中散射介质状态不固定,目标结构具有较高复杂度以及可获取的训练散射数据有限的情况下,单纯利用数据驱动的方法已无法进行准确高效的重建。将散斑相关原理和卷积神经网络强大的数据挖掘和映射能力进行有效的结合,进一步挖掘和利用散斑所包含的冗余信息,实现了仅利用一块薄散射介质对应的散斑数据即可实现透过具有不同统计特性散射介质的复杂目标重构。该方法针对实际散射场景复杂多变和训练样本数据有限的情况,实现了对复杂目标的高质量恢复,有力地推动了基于物理感知的学习方法在实际散射场景中的应用。     

Single image shadow detection via uncertainty analysis and GCN-based refinement strategy

Learning-based shadow detection methods have achieved an impressive performance, while these works still struggle on complex scenes, especially ambiguous soft shadows. To tackle this issue, this work proposes an efficient shadow detection network (ESDNet) and then applies uncertainty analysis and graph convolutional networks for detection refinement. Specifically, we first aggregate global information from high-level features and harvest shadow details in low-level features for obtaining an initial prediction. Secondly, we analyze the uncertainty of our ESDNet for an input shadow image and then take its intensity, expectation, and entropy into account to formulate a semi-supervised graph learning problem. Finally, we solve this problem by training a graph convolution network to obtain the refined detection result for every training image. To evaluate our method, we conduct extensive experiments on several benchmark datasets, i.e., SBU, UCF, ISTD, and even on soft shadow scenes. Experimental results demonstrate that our strategy can improve shadow detection performance by suppressing the uncertainties of false positive and false negative regions, achieving state-of-the-art results.     

Semantic guided knowledge graph for large-scale zero-shot learning

Zero-shot learning has received growing attention, which aims to improve generalization to unseen concepts. The key challenge in zero-shot tasks is to precisely model the relationship between seen and unseen classes. Most existing zero-shot learning methods capture inter-class relationships via a shared embedding space, leading to inadequate use of relationships and poor performance. Recently, knowledge graph-based methods have emerged as a new trend of zero-shot learning. These methods use a knowledge graph to accurately model the inter-class relationships. However, the currently dominant method for zero-shot learning directly extracts the fixed connection from off-the-shelf WordNet, which will inherit the inherent noise in WordNet. In this paper, we propose a novel method that adopts class-level semantic information as a guidance to construct a new semantic guided knowledge graph (SG-KG), which can correct the errors in the existing knowledge graph and accurately model the inter-class relationships. Specifically, our method includes two main steps: noise suppression and semantic enhancement. Noise suppression is used to eliminate noise edges in the knowledge graph, and semantic enhancement is used to connect two classes with strong relations. To promote high efficient information propagation among classes, we develop a novel multi-granularity fusion network (MGFN) that integrates discriminative information from multiple GCN branches. Extensive experiments on the large-scale ImageNet-21K dataset and AWA2 dataset demonstrate that our method consistently surpasses existing methods and achieves a new state-of-the-art result.     

Deep Convolutional Neural Network Approach for COVID-19 Detection

Coronavirus disease 2019 (Covid-19) is a life-threatening infectious disease caused by a newly discovered strain of the coronaviruses. As by the end of 2020, Covid-19 is still not fully understood, but like other similar viruses, the main mode of transmission or spread is believed to be through droplets from coughs and sneezes of infected persons. The accurate detection of Covid-19 cases poses some questions to scientists and physicians. The two main kinds of tests available for Covid-19 are viral tests, which tells you whether you are currently infected and antibody test, which tells if you had been infected previously. Routine Covid-19 test can take up to 2 days to complete; in reducing chances of false negative results, serial testing is used. Medical image processing by means of using Chest X-ray images and Computed Tomography (CT) can help radiologists detect the virus. This imaging approach can detect certain characteristic changes in the lung associated with Covid-19. In this paper, a deep learning model or technique based on the Convolutional Neural Network is proposed to improve the accuracy and precisely detect Covid-19 from Chest Xray scans by identifying structural abnormalities in scans or X-ray images. The entire model proposed is categorized into three stages: dataset, data pre-processing and final stage being training and classification.     

基于双重注意力机制和GRU网络的短期负荷预测模型

电力负荷预测对电力系统的部署、规划和运行影响重大,但目前各输入特征对电网负荷情况影响的程度不稳定,且递归神经网络捕获负荷数据的长期记忆能力差,导致预测精度下降。提出一种基于双重注意力机制和GRU网络的预测新模型,利用特征注意力机制自主分析历史信息与输入特征间的关联关系,提取重要特征,并通过时序注意力机制自主选取GRU网络中关键时间点的历史信息,提升较长时间段预测效果的稳定性。在3个公开数据集上的实验结果表明,该模型在预测精度指标上表现良好,对比SVR、KPCA-ELM、DBN、GRU、Attention-GRU、CNN-LSTM、Attention-CNN-GRU模型预测精度分别提高了2.47、1.14、1.93、1.37、1.04、0.74、0.41个百分点。     

基于因果干预与不变性的卷积预训练模型优化研究

基于卷积神经网络（CNN）的深度模型在图像识别与分类领域应用广泛,但在全局特征控制、概念层次特征不变性提取和变量之间的因果关系确定方面仍存在不足,使得深度模型缺乏灵活性、适应性及泛化性。基于因果干预和不变性,提出一种基于CNN深度模型的定向修剪和网络结构优化方法。通过对模型输入进行基于不变性的干预调制,根据生成的调制图片序列分析预训练网络卷积子结构的输出分布,筛选和定向修剪噪声敏感子结构。构建基于类间区分度的目标函数,借助经济学领域中的资本资产定价模型构建网络的层间连接,生成在单分类任务下能增大类间区分度的网络拓扑结构,逐层优化构建概念层次的稳定特征。在ImageNet-2012数据集上的实验结果表明,优化后的深度模型相比于ResNet50基线预训练模型的分类准确率约提升了5个百分点,并大幅降低了训练集规模。