A study of the “heartbeat spectra” for “sleeping beauties”

We first introduced interesting definitions of “heartbeat” and “heartbeat spectrum” for “sleeping beauties”, based on van Raan's variables. Then, we investigated 58,963 papers of Nobel laureates during 1900–2000 and found 758 sleeping beauties. By proposing and using G_s index, an adjustment of Gini coefficient, to measure the inequality of “heartbeat spectrum”, we observed that publications which possess “late heartbeats” (most citations were received in the second half of sleeping period) have higher awakening probability than those have “early heartbeats” (most citations were received in the first half of sleeping period). The awakening probability appears the highest if an article's G_s index exists in the interval [0.2, 0.6).     

偶数元择多函数密码学性质的研究

择多函数是一类具有最优代数免疫的布尔函数,受到研究者的广泛关注.本文主要讨论了偶数元择多函数的代数次数,非线性度,相关免疫性,并利用偶数元择多函数的Walsh谱特性,构造出了与其线性等价的相关免疫函数.     

基于显著区域纹理谱特征的图像检索方法

提出一种基于显著区域纹理谱特征的图像检索方法.首先计算像素点显著度,生成显著图;然后对图像的各个分割区域,按照显著图确定显著区域;再提取各个显著区域的纹理谱直方图特征;最后给出显著区域纹理谱特征的加权函数.实验结果表明,该方法能将视觉敏感度与纹理特征有效地结合起来,在不增加太多计算时间的条件下,提高检索质量.     

光在ZnO二维无序介质中的光学特性研究

基于ZnO二维无序介质的三角形排列结构模型,采用时域有限差分法研究了光在ZnO二维无序介质中的光学特性.结果表明,无序介质中光场空间分布呈现局域化特征.计算了局域化较强处的透射谱,发现探测点在ZnO增益范围内存在多个透射共振峰.讨论了共振峰对应的局域模的光场分布,得到不同局域模光场具有不同的空间分布.     

两种Co(Ⅲ)配合物键合异构体的制备及结构分析

本文通过条件选择合成两种Co(Ⅲ)配合物键合异构体[Co(NH3)5NO2]Cl2(配合物1)与[Co(NH3)5ONO]Cl2(配合物2),通过红外光谱、紫外光谱对配合物进行表征。确立两种键合异构体可能的结构。     

基于神经网络的车联网频谱感知组合算法

针对车联网环境下无线频谱资源短缺的问题,提出一种基于神经网络的多条件频谱感知组合算法.该算法利用神经网络较强的多分类能力,将信号能量、协方差矩阵的最大特征值、最小特征值、迹和平均特征值融合作为神经网络特征参数实现合作频谱感知,并从理论上分析参数选择方案,算法还充分考虑信道多径衰落和阴影效应导致的信噪比很低的情况以及车辆...     

Engaging with animal-assisted interventions (AAIs): exploring the experiences of young people with ASD/ADHD diagnoses

Animal-assisted Interventions (AAIs) have been used as therapeutic interventions aimed at improving psychological well-being, often for young people with mental health and educational difficulties. This qualitative study explored how three students (male and female), aged 12–15 and with ASD and/or ADHD diagnoses experienced AAI at an alternative education provision. Semi-structured interviews with each participant were conducted and analysed using constructivist thematic analysis. Three themes and four sub-themes were identified. The themes were: (1) Self-esteem, with sub-themes motivation and reward, and self-awareness, (2) Emotional benefits with strategy building, and support and (3) Identification. Participants valued AAI as a positive intervention which aided their psychological well-being. Findings emphasise the effectiveness of AAIs for young people with neurodevelopmental disorders in both a therapeutic and an educational context.     

基于卷积编解码网络的自由感应衰减信号提取方法北大核心

在实际应用中,自由感应衰减信号(FID)数据中通常含有大量噪声干扰,直接影响了检测结果的准确性。该文针对上述问题,设计了一种基于卷积编解码网络的FID信号提取方法。即在卷积神经网络的框架下,通过训练学习复杂噪声背景下FID数据时频谱与纯净FID信号时频谱之间的映射关系,实现FID信号的噪声压制,再经短时傅里叶逆变换完成FID信号的提取。在卷积网络训练中同时考虑了FID信号的实部和虚部特征,更好地保留了其相位信息。实验结果表明:提出的FID信号提取方法可以在多种噪声同时存在的情况下有效地压制噪声,且不损失FID信号。该方法可作为深度学习课程的验证性实验,也可作为创新性实验内容。