基于注意力机制的GRU神经网络安全态势预测方法

传统的网络安全态势预测方法依赖于历史态势值的准确性,并且各种网络安全因素之间存在相关性和重要程度差异性。针对以上问题,提出一种基于注意力机制的循环门控单元(recurrent gate unit, GRU)编码预测方法,该方法利用GRU神经网络挖掘网络安全态势数据之间的时间相关性;引入注意力机制计算安全指标的分配权重并将其编码为网络安全态势值;利用改进的粒子群优化(particle swarm optimization, PSO)算法进行超参数寻优,以加速GRU神经网络的训练。仿真分析表明,所提方法具有更快的收敛速度和较低的复杂度,并且在不同的预测时长下具有较小的均方误差和平均绝对误差。     

几种诱导昆虫细胞RNA干扰质粒的构建和效果比较

构建了四种可以在昆虫细胞中表达形成dsRNA的质粒,比较了其诱导RNA干扰,抑制目标基因———绿色荧光蛋白(GFP)基因表达的效果.四种质粒都不同程度地抑制了质粒介导的GFP表达,其中尤以单个果蝇hsp70启动子表达反向重复序列,并带有杆状病毒AcMNPV增强子hr5的质粒效果最佳,使GFP表达量下降到原来的3.9%.当用于抑制由杆状病毒介导的GFP表达时,以两个相向排列的AcMNPV ie1基因启动子的构造有明显的抑制效果.这些结果对于在昆虫细胞中有效地利用RNA i研究基因功能有参考意义.     

可重构视频阵列处理器测试平台设计与实现

针对可重构视频阵列处理器的设计要求及传统测试方法测试视频编解码系统时速度慢、精度低和可观测性不强的问题。开发了基于Qt的用户界面,设计实现了以现场可编程门阵列(Field programmable gate-array,FPGA)为核心的软硬件协同测试平台。在PC端实现以软件仿真为基础的数据传输与图像重现,在FPGA端实现以可重构视频阵列处理器为基础的视频编解码算法并行映射。实验结果表明,在工作频率为100 MHz时,FPGA与PC之间可正确传输数据并满足算法测试时不同测试用例的更换需求,具有较好的可观测性。     

基于LWSVR的繁忙机场航班滑出时间预测

针对繁忙机场航班滑出时间预测准确率低的问题,结合局部回归和加权支持向量回归,提出基于局部加权支持向量回归的离港航班滑出时间预测模型。该模型采用K最近邻方法,减小训练样本集容量,并为每个预测样本构建一个预测模型。通过计算训练样本与预测样本间的马氏距离,来优化加权支持向量回归中高斯核加权函数的带宽参数,获得加权系数。结合某机场离港航班数据仿真分析,实验结果表明模型在误差允许范围内的预测准确率达到83.33%,模型更加稳定。     

控制污染物排放的支线机场跑滑结构优化仿真

民航运输在LTO(Landing and take-off)循环阶段的大气污染物排放对地面人员和近地环境危害较大,应主要通过减少地面滑行时间进行控制。考虑支线机场常见的单跑滑结构给出飞机地面滑行时间计算方法,基于该方法提出了污染物评价指标并建立了支线机场航空器污染物排放评估模型。以某支线机场为例,使用Matlab建模计算得出污染物评价指数最小滑行道口位置。利用AirTOp仿真验证了模型计算结果有效性,优化后的跑道出口位置将使该机场污染物总指数较原位置降低21.7%。     

时间反演联合接收端人工噪声的安全传输方案

针对时间反演(time reversal, TR)技术无法较好地解决聚焦区内非法用户窃听的问题,提出了一种TR技术联合接收端人工噪声的物理层安全传输方案。首先,利用时间反演技术的空时聚焦性提高合法用户接收端的信噪比(signal to noise ratio, SNR)。其次,通过在接收端加入人工噪声来干扰非法用户对合法用户保密信息的窃听,以聚焦区内外系统安全性能为基础,推导出保密信干噪比(signal to interference plus noise ratio, SINR)、可达保密速率和误码率(bit error rate, BER)的解析表达式。最后,通过仿真分析证明了所提方案能提高系统的SINR和可达保密速率,同时还能降低合法用户的BER,使系统拥有更好的安全性能。     

A novel fully-integrated miniature six-axis force/torque sensor

This paper presents a new designed miniature six DOF （degree of freedom） force/torque sensor. This sensor is fully integrated with a micro DSP （digital signal processor）, so all the signal conditioning, A/D, decoupling, digital-signals serial output are performed in the sensor. Some experimental results are presented to demonstrate the capability of the proposed design. Finally, a neural network was used for decoupling the interacting signals, compared with the conventional method using the inverse matrix, this new method is more accurate.     

Cystic fibrosis transmembrane conductance regulator—emerging regulator of cancer

Mutations of cystic fibrosis transmembrane conductance regulator (CFTR) cause cystic fibrosis, the most common life-limiting recessive genetic disease among Caucasians. CFTR mutations have also been linked to increased risk of various cancers but remained controversial for a long time. Recent studies have begun to reveal that CFTR is not merely an ion channel but also an important regulator of cancer development and progression with multiple signaling pathways identified. In this review, we will first present clinical findings showing the correlation of genetic mutations or aberrant expression of CFTR with cancer incidence in multiple cancers. We will then focus on the roles of CFTR in fundamental cellular processes including transformation, survival, proliferation, migration, invasion and epithelial–mesenchymal transition in cancer cells, highlighting the signaling pathways involved. Finally, the association of CFTR expression levels with patient prognosis, and the potential of CFTR as a cancer prognosis indicator in human malignancies will be discussed.     

A genome-wide association study in Chinese men identifies three risk loci for non-obstructive azoospermia

Non-obstructive azoospermia (NOA) is one of the most severe forms of male infertility. Its pathophysiology is largely unknown, and few genetic influences have been defined. To identify common variants contributing to NOA in Han Chinese men, we performed a three-stage genome-wide association study of 2,927 individuals with NOA and 5,734 controls. The combined analyses identified significant (P < 5.0 × 10(-8)) associations between NOA risk and common variants near PRMT6 (rs12097821 at 1p13.3: odds ratio (OR) = 1.25, P = 5.7 × 10(-10)), PEX10 (rs2477686 at 1p36.32: OR = 1.39, P = 5.7 × 10(-12)) and SOX5 (rs10842262 at 12p12.1: OR = 1.23, P = 2.3 × 10(-9)). These findings implicate genetic variants at 1p13.3, 1p36.32 and 12p12.1 in the etiology of NOA in Han Chinese men.