A Novel Training Sequence Applied to DCS-Based Channel Estimation

Studies have indicated that the distributed compressed sensing based(DCSbased) channel estimation can decrease the length of the reference signals effectively. In block transmission, a unique word(UW) can be used as a cyclic prefix and reference signal. However, the DCS-based channel estimation requires diversity sequences instead of UW. In this paper, we proposed a novel method that employs a training sequence(TS) whose duration time is slightly longer than the maximum delay spread time. Based on proposed TS, the DCS approach perform perfectly in multipath channel estimation. Meanwhile, a cyclic prefix construct could be formed, which reduces the complexity of the frequency domain equalization(FDE) directly. Simulation results demonstrate that, by using the method of simultaneous orthogonal matching pursuit(SOMP), the required channel overhead has been reduced thanks to the proposed TS.

Optimal training sequences for MIMO systems under correlated fading

The optimal design of training sequences for channel estimation in multiple-input multiple-output （MIMO） systems under spatially correlated fading is considered. The channel is assumed to be a block-fading model with spatial correlation known at both the transmitter and the receiver. To minimize the channel estimation error, optimal training sequences are designed to exploit full information of the spatial correlation under the criterion of minimum mean square error （MMSE）. It is investigated that the spatial correlation is helpful to decrease the estimation error and the proposed training sequences have good performance via simulations.

Time-domain training sequences design for MIMO OFDM channel estimation

This paper describes a Least Squares (LS) channel estimation scheme for MIMO OFDM systems based on time-domain training sequence. We first compute the minimum mean square error (MSE) of the LS channel estimation, and then derive the optimal criteria of the training sequence with respect to the minimum MSE. It is shown that optimal time-domain training sequence should satisfy two criteria. First, the autocorrelation of the sequence transmitted from the same antenna is an impulse function in a region longer than the channel maximum delay. Second, the cross-correlation between sequences transmitted from different antennas is zero in this region. Simulation results show that the estimator using optimal time-domain training sequences has better performance than that using optimal frequency training sequence at low signal-to-noise ratio (SNR). To reduce the training overhead, a suboptimal training sequence is also proposed. Comparing with optimal training sequence, it has low computation complexity and high transmission efficiency at the expense of little performance degradation.

An efficient training sequences strategy for channel estimation in OFDM systems with transmit diversity

This paper deals with channel estimation for orthogonal frequency-division multiplexing (OFDM) systems with transmit diversity. Space time coded OFDM systems, which can provide transmit diversity, require perfect channel estimation to improve communication quality. In actual OFDM systems, training sequences are usually used for channel estimation. The authors propose a training based channel estimation strategy suitable for space time coded OFDM systems. This novel strategy provides enhanced performance, high spectrum efficiency and relatively low computation complexity.

ON CHANNEL ESTIMATION USING OPTIMAL TRAINING SEQUENCES IN CYCLIC-PREFIX-BASED SINGLE-CARRIER SYSTEMS WITH SPACE-TIME BLOCK-CODING

In this paper, a new scheme that combines Space-Time Block-Coding (STBC) based on an Alamouti-like scheme and the Least Squares (LS) channel estimation using optimal training sequences in Cyclic-Prefix-based (CP)\Single-Carrier (SC) systems is proposed. With two transmit antennas, based on Cramer-Rao lower bound for channel estimation, it is shown that the Periodic Comple- mentary Set (PCS) is optimal over frequency-selective fading channels. Compared with the normal scheme without STBC, 3dB Mean Square Error (MSE) performance gains and fewer restrictions on the length of channel impulse response are demonstrated.

Training Sequence Design for Coordination on Multiple Point System

Although the Cramer-Rao Bound(CRB) can be used as the benchmark of estimation algorithm performance,it's too complicated for joint training sequence(TS) design for multiple input multiple output(MIMO) orthogonal frequency division multiplexing(OFDM) coordination on multiple point(CoMP) systems.So a minimum mean square error(MSE) based sub-optimal sequence design criterion was proposed,including ideal sequence correlation property and sequence length constraint.The simulation results verify the theory analysis.

System Restoration Tools: System Restoration Navigator Integrated into EPRI Operator Training Simulator (SRN/OTS)

System operators and planners develop and implement restoration plans based on off-line simulation studies, and accumulated experience and knowledge. One of the challenges in developing a restoration plan is to sift through numerous possible restoration scenarios and paths, in order to identify those that are technically feasible. When implementing a restoration plan in an on-line environment following a blackout, the operators need to adapt to the actual outage scenarios and available resources, and be constantly mindful of anticipated voltage and frequency excursions that must remain within system and equipment tolerances. In recognition of these challenges, EPRI has developed System Restoration Navigator (SRN), to provide decision support to system restoration planning and operations engineers in developing, evaluating and revising system restoration strategies, guidelines, plans and step-by-step procedures. During 2013-2014, EPRI developed SRN version 3.0, which is designed to facilitate its integration into a commercial operator training simulator (OTS) (AKA a dispatch training simulator, DTS). The integration of SRN 3.0 with an OTS allows operators to obtain experience in simulating, developing, experimenting with and revising system restoration plans, and to address related regulatory standards. The integration expands the usability of SRN 3.0 by providing the OTS platform for training purposes and for the purpose of interfacing SRN 3.0 with operational power system models to be able to explore near real time application of SRN 3.0. This 2013-2014 development work also included the integration of SRN 3.0 into EPRI OTS, and its application on the Florida Reliability Coordinating Council (FRCC) power system. A detailed account of development of SRN 3.0, its integration into EPRI OTS and its application to FRCC system is presented in this paper.

A PTS Optimization Scheme with Superimposed Training for PAPR Reduction in OFDM System

Partial Transmit Sequences (PTS) is an efficient scheme for Peak-to-Average Power Ratio (PAPR) reduction in Orthogonal Frequency Division Multiplexing (OFDM) system. It does not bring any signal distortion. However, its remarkable drawback is the high computational complexity. In order to reduce the computational complexity, currently many PTS methods have been proposed but with the cost of the loss of PAPR performance of the system. In this paper, we introduce an improved PTS optimization method with superimposed training. Simulation results show that, compared with conventional PTS, improved PTS scheme can achieve better PAPR performance while be implemented with lower computation complexity of the system.

Virtual reality simulators for gastrointestinal endoscopy training

The use of simulators as educational tools for medical procedures is spreading rapidly and many efforts have been made for their implementation in gastrointestinal endoscopy training. Endoscopy simulation training has been suggested for ascertaining patient safety while positively influencing the trainees' learning curve. Virtual simulators are the most promising tool among all available types of simulators. These integrated modalities offer a human-like endoscopy experience by combining virtual images of the gastrointestinal tract and haptic realism with using a customized endoscope. From their first steps in the 1980s until today, research involving virtual endoscopic simulators can be divided in two categories: investigation of the impact of virtual simulator training in acquiring endoscopy skills and measuring competence. Emphasis should also be given to the financial impact of their implementation in endoscopy, including the cost of these state-of-theart simulators and the potential economic benefits from their usage. Advances in technology will contribute to the upgrade of existing models and the development of new ones; while further research should be carried out to discover new fields of application.

Diffusion tensor imaging of spinal microstructure in healthy adults： improved resolution with the readout segmentation of long variable echo-trains

Diffusion tensor imaging plays an important role in the accurate diagnosis and prognosis of spinal cord diseases. However, because of technical limitations, the imaging sequences used in this technique cannot reveal the fine structure of the spinal cord with precision. We used the readout segmentation of long variable echo-trains（RESOLVE） sequence in this cross-sectional study of 45 healthy volunteers aged 20 to 63 years. We found that the RESOLVE sequence significantly increased the resolution of the diffusion images and improved the median signal-to-noise ratio of the middle（C4–6） and lower（C7–T1） cervical segments to the level of the upper cervical segment. In addition, the values of fractional anisotropy and radial diffusivity were significantly higher in white matter than in gray matter. Our study verified that the RESOLVE sequence could improve resolution of diffusion tensor imaging in clinical applications and provide accurate baseline data for the diagnosis and treatment of cervical spinal cord diseases.

基于生成对抗Transformer的电力负荷数据异常检测

电力负荷异常数据将给电力系统规划、负荷预测以及用能分析等带来较大的负面影响,因此亟须对负荷数据异常进行检测与识别。首先,针对电力负荷数据异常分类、原因及其特征开展分析。其次,改进传统Transformer编码器结构,采用多头注意力层代替掩码多头注意力层,同时移除前馈网络,以提高模型对负荷时序序列的全局注意力。基于生成对抗网络(generative adversarial networks,GAN)生成器与判别器的博弈结构,提出一种改进的GAN-Transformer模型,以更好地捕捉趋势性特征并加速模型收敛。然后,引入多阶段映射与训练方法,综合焦点分数打分机制,通过分阶段负荷序列重构帮助模型更好地提取负荷数据异常特征。最后,算例分析结果表明,GAN-Transformer模型在负荷数据异常检测精确率、召回率、F_(1)值以及训练时间方面均具有更优的性能,验证了所提方法的有效性和优越性。文中研究工作为基于深度学习进一步实现电力负荷数据异常分类与数据修复提供了有益参考。

基于Ts-EVI时间序列谱的冬小麦面积提取

冬小麦是中国重要的粮食作物,利用遥感技术准确获取冬小麦分布面积对于从宏观上实现冬小麦生长状况和病虫害监测预警等田间精细管理以及产量预测具有十分重要的意义。采用NASA提供的2005年2～6月MYD09A1和MYD11A2数据产品,结合冬小麦生育期形态变化特征,以Ts和EVI构建的时间序列谱对河北省冬小麦种植面积的提取方法进行了研究。结果表明:冬小麦在Ts-EVI特征空间中具有独特的序列谱相特征,返青期后冬小麦的Ts在整体上表现为逐渐升高,EVI在开花期后下降,灌浆后期快速降低;对Ts-EVI空间信息的一维化处理后,采用主成分分析方法对时间信息降维,以ISODATA非监督分类方法实现冬小麦分离提取结果的样点验证精度为91.20%;Ts-EVI时间序列谱反映的作物生长过程的生理意义明确,采用Ts-EVI时间序列谱的遥感分类方法可以较好地实现冬小麦的遥感分类提取,具有较好的可行性。

融入时间信息的预训练序列推荐方法

序列推荐旨在根据用户与项目的历史交互序列,学习用户动态偏好,为用户推荐后续可能感兴趣的项目。基于预训练模型在适应下游任务方面具有优势,预训练机制在序列推荐中备受关注。现有序列推荐预训练方法忽略了现实中时间对用户交互行为的影响,为了更好地捕获用户与项目交互的时间语义,提出了融入时间信息的预训练序列推荐模型TPTS-Rec(Time-aware Pre-Training method for Sequence Recommendation)。首先,在嵌入层引入时间嵌入矩阵以获取用户交互项目与时间的关联信息。然后,在自注意力层采用同一时间点采样的方法以学习项目间的时间关联信息。最后,在微调阶段从时间维度扩增用户交互序列长度以缓解数据稀疏性问题。在真实数据集上的对比实验结果表明,与基线模型相比,所提模型TPTS-Rec的推荐效果有显著提升。

基于高斯似然的精准水声信道估计

针对时变水声信道的多途干扰问题,提出基于高斯似然(Gaussian likelihood,GL)的精准水声信道估计算法。GL算法将相邻信道短块的高斯概率密度函数相乘,乘积仍然服从高斯分布,且方差变小,从而进一步提高信道估计的准确性;采用叠加训练(superimposed training,ST)方案,将训练序列和符号序列线性叠加,使训练序列持续传输,实现对信道的实时跟踪。将ST方案、GL算法和Turbo均衡以迭代的方式相结合,估计出的符号序列作为虚拟训练序列,进一步提高时变水声信道的估计和跟踪性能。通过多次迭代计算,实现时变水声信道的精准估计和实时跟踪。最后,通过计算机仿真以及胶州湾收发节点水平距离500 m和5.5 km的海上运动实装试验,验证了所提算法的有效性。

基于工况序列寻优的列车节能操纵策略优化

传统“三阶段”“四阶段”工况序列的列车操纵模式难以满足城市轨道交通复杂线路条件下的列车节能运行需求,本文提出一种基于工况序列寻优的列车节能操纵策略优化方法。将物理区间离散为多个等距离子区间,建立工况序列与子区间之间的映射关系,以区间总牵引能耗和运行时间最小化为目标构建列车节能操纵策略多目标优化模型。为提高模型求解效率,改进非支配排序遗传算法的交叉算子和距离算子,通过离散仿真求解列车节能操纵策略集。以福州地铁1号线的两个典型区间为对象进行案例分析,结果表明,相比传统操纵模式,优化后列车牵引能耗平均降低约19%。本文方法通过合理选择算法参数能够有效构建适应不同线路条件的运行工况序列,生成秒级运行时间划分下的列车节能操纵策略集。

基于API序列和预训练模型的恶意软件检测

针对现有方法存在特征表达受限、无法捕获API序列全局语义信息,且恶意软件数据集通常包含大量无标注数据,无法直接进行有监督学习等问题,利用自然语言预训练模型技术,提出一种基于API调用序列和预训练模型的恶意软件检测方法。使用原始API序列构建分词器;基于BERT模型构建出动态掩码序列模型进行无监督学习的预训练,同时获取API序列的全局动态编码表示;使用该编码构造检测模型。实验结果表明,所提方法能有效检测出恶意软件。

基于扩散序列的多元可控文本生成

随着大规模预训练语言模型的出现,文本生成技术已取得突破性进展。然而,在开放性文本生成领域,生成的内容缺乏拟人化的情感特征,使生成的文本难以让人产生共鸣和情感上的联系,可控文本生成在弥补当前文本生成技术不足方面具有重要意义。首先,在ChnSentiCorp数据集的基础上完成主题和情感属性的扩展,同时,为构建一个可生成流畅文本且情感丰富的多元可控文本生成模型,提出一种基于扩散序列的可控文本生成模型DiffuSeq-PT。该模型以扩散模型为基础架构,利用主题情感属性和文本数据在无分类器引导条件下对序列执行扩散过程,使用预训练模型ERNIE 3.0(Large-scale Knowledge Enhanced Pre-training for Language Understanding and Generation)的编码解码能力贴合扩散模型的加噪去噪过程,最终生成符合相关主题和多情感粒度的目标文本。与基准模型DiffuSeq相比,所提模型在2个公开的真实数据集(ChnSentiCorp和辩论数据集)上分别取得0.13和0.01的BERTScore值的提升,困惑度分别下降了14.318和9.46。

不同叠加训练序列PTS降低PAPR性能比较

部分传输序列(PTS)能有效降低正交频分复用(OFDM)系统的峰均功率比(PAPR),但是其存在降低性能有限和边带信息占用频谱资源的问题。文中在传统PTS方法的基础上,基于叠加训练序列PTS降低PAPR的方法,分析了叠加不同训练序列、不同训练序列作相位旋转因子及功率分配因子对降低PAPR性能的影响。仿真结果表明:叠加自相关性较好的训练序列,降低PAPR的能力更强;不同训练序列作相位旋转因子时,对降低PAPR的效果影响不大;采用Hadama码、功率分配因子大于0.15时,降低PAPR的能力增强。该方法中的训练序列,不仅可以用于同步﹑信道估计,而且可以降低PAPR,并提高系统频谱利用率。

基于CRCCTst的CTCS-3级列控系统测试方法研究

依据CTCS-3级列控系统技术规范,研究基于CRCCTst的CTCS-3级列控系统测试方法。以《CTCS-3级列控车载设备补充技术规范(暂行)》为例,选取部分功能需求为依据,以ATP车载设备为被测对象,编写符合规范场景的测试案例并运行测试序列,对被测设备进行测试,验证了基于CRCCTst的CTCS-3级列控系统测试方法的可用性。

基于XLNET模型的开阳磷矿成矿条件相关地质实体识别与应用

随着磷矿找矿难度越来越大,地质勘探成果报告也愈来愈多,通过人工识别海量文档中与磷矿成矿相关地质信息耗时低效,无法满足知识共享传播和地质报告智能管理的需求。为快速获得磷矿地质文档报告中隐藏的成矿地质知识,基于XLNET模型建立了磷矿成矿地质实体自动识别的方法。首先对实体进行BIO标注建立地质实体字典,利用XLNET作为底层预处理模型学习句子双向语义;然后使用BILSTM-Attention-CRF模型实现文本多标签的智能分类;最后通过定位磷矿实体在报告中的分布位置大致推测该处磷矿成矿条件和成矿模式。将该模型与其余3种模型比较得出结果,该模型识别的准确率(P)、召回率(R)及F1值都接近了90%,较前3种模型分别调高了2%,5%,6%。该研究为开阳磷矿地质研究人员提供了更加高效的地质实体自动识别的方法。