The application of three-dimensional seismic spectral decomposition and semblance attribute to characterizing the deepwater channel depositional elements in the Taranaki Basin of New Zealand

In the past few years, three-dimensional（3-D） seismogram has become an essential tool for the interpretation of subsurface stratigraphy and depositional systems. Seismic stratigraphy in conjunction with seismic geomorphology has elevated the degree to which seismic data can facilitate geological interpretation, especially in a deepwater environment. Technologies such as time slicing and interval attribute analysis can enhance geomorphological interpretations, and, when integrated with stratigraphic analyses, can yield insights regarding distribution of seal and reservoir facies. Multiple attributes corendering can further bring out features of geological interest that other technologies may overlook. This method involves corender spectral decomposition components（SDC） with semblance attributes to describe the distribution of deepwater channel elements and the boundaries of deepwater sinuous channel. Applying this technology to four elements is observed：（1） point-bars,（2） migration of channel meander loops,（3） channel erosion/cut, and（4） avulsion. The planview expression of the deepwater channel ranges from low sinuosity to high sinuosity. Furthermore, this technology has enabled interpreters to visualize details of complex depositional elements and can be used to predict net-to-gross ratio in channel systems, which can be incorporated into borehole planning for exploration as well as development needs to improve risk management significantly. The technology is applied to the study area in an effort to illustrate the variety of interpretation technologies available to the geoscientist.

Spectral Correction Method of Multi-Channels Near-Infrared Spectrometer and Applications

Near-infrared (NIR) spectrometer based on semiconductor lasers can combine light source and splitter into one, which is an important direction for development of miniature instruments. In order to avoid random interference caused by inconsistency between light sources, the novel evaluation indicators for global stability of multi-channels spectral system were proposed based on the correlation between dynamic deviation spectra of any two channels. The NIR analysis of moisture for corn powder samples based on the partial least squares combined with Savitzky-Golay (SG) smoothing was taken as an example, and a spectral correction method for enhancing prediction performance of multi-channels spectral system was further provided using above evaluation indicators. The experiment results showed that the global stability evaluation indicators significantly increased after SG smoothing correction. Meanwhile, the root-mean-square errors of prediction for corn moisture reduced from 0.373 to 0.283 (%), and the correlation coefficient between predicted and actual values was improved from 0.702 to 0.855. The above results indicated that by improving global stability indicators, the prediction ability of multi-channels spectral system can be improved. The proposed method provided a valuable reference for designing multi-channels diminutive spectrometer with high prediction performance, which had significance for large-scale application of NIR technology.

Single-Channel Speech Enhancement Using Critical-Band Rate Scale Based Improved Multi-Band Spectral Subtraction

This paper addresses the problem of single-channel speech enhancement in the adverse environment. The critical-band rate scale based on improved multi-band spectral subtraction is investigated in this study for enhancement of single-channel speech. In this work, the whole speech spectrum is divided into different non-uniformly spaced frequency bands in accordance with the critical-band rate scale of the psycho-acoustic model and the spectral over-subtraction is carried-out separately in each band. In addition, for the estimation of the noise from each band, the adaptive noise estimation approach is used and does not require explicit speech silence detection. The noise is estimated and updated by adaptively smoothing the noisy signal power in each band. The smoothing parameter is controlled by a-posteriori signal-to-noise ratio (SNR). For the performance analysis of the proposed algorithm, the objective measures, such as, SNR, segmental SNR, and perceptual evaluations of the speech quality are conducted for the variety of noises at different levels of SNRs. The speech spectrogram and objective evaluations of the proposed algorithm are compared with other standard speech enhancement algorithms and proved that the musical structure of the remnant noise and background noise is better suppressed by the proposed algorithm.

顾及信号振荡特征的慢滑移信息时空提取

针对传统滤波和固定函数拟合等方法探测慢滑移信号时易产生信息误剔除或伪信号提取等问题,提出一种基于信号振荡特征的慢滑移时空信息探测方法:利用多通道奇异谱分析分解坐标序列;然后根据慢滑移位移特征确定信号的起止时间;最后基于信号的振荡方向和振幅归一化明确空间响应方向和强度。仿真结果表明,相较于主成分分析和独立成分分析,该方法在探测慢滑移起止时间、空间响应方向和强度等方面优势明显:实际应用中,利用该方法成功探测出新西兰马纳瓦图的慢滑移事件,测站WANG和PNUI连线两侧的空间响应方向相反,强度大,地震危险性高;而独立成分分析和主成分分析探测到的空间响应可能受到共模误差影响,时间响应也无法明确慢滑移起止时间。

多通道抗串扰式光谱共焦位移测量方法

提出一种位移测量方法,解决扫描光谱共焦测量系统存在不同共焦点之间的相互干扰问题。通过光纤束保持多个测量通道之间的独立性,并利用光开关切换测量通道来防止干扰。该系统的7个共焦点构成一个等边六边形,其中6个点分别位于六边形的各顶点处,1个共焦点位于六边形的中心,六边形的边长为43.27μm。通过位移测量实验得出,位移测量系统各测量通道的绝对误差小于0.15μm,范围为0~90μm。在测量台阶高度的实验中,2个高度差为30μm的台阶面,误差最大者为0.16μm;在球径测量实验中,无需移动测量系统或被测物体便可完成7路通道的测量;通过对数据进行拟合得到半径为5 mm的球体近似值,最大误差为80.92μm。对比传统方法,该方法展现了较高的精度和分辨率。

结合区域引导和双注意力机制的高光谱目标检测判别式学习网络

高光谱图像(HyperSpectral Images,HSIs)具有高光谱分辨率和丰富的光谱信息,其具有的大量窄波段电磁波有利于获取感兴趣目标的理化信息,并根据对应的光谱特征对不同物质进行有效区分,从而完成目标检测任务.然而有限样本、少量先验信息、高维相似背景及不同类别差异小所导致的目标和背景混淆问题使得高光谱目标检测(Hyperspectral Target Detection,HTD)面临挑战.为此,本文提出结合区域引导和双注意力机制的高光谱目标检测判别式学习网络(Region-guided and dual-Attention Discriminative learning Network,RADN),以缓解标记样本少的条件下不同类别相似度高和相同类别差异性大导致的背景和目标不易区分的问题,减少高维冗余特征带来的计算复杂度,同时提升检测精度.本文使用经验性区域引导网络训练,采用光谱约束的无监督聚类方法确定网络输入,选择性地关注高光谱图像中的显著性特征和感兴趣区域.此外,本文在网络中添加双通道注意力机制来辅助复杂背景分布的估计,并在网络中引入不同类别光谱先验损失函数,进一步减少高维复杂背景以及光谱变化对于目标的干扰.实验结果和分析表明,RADN在不同数据集上的性能优于现有先进的算法.

协同全极化SAR与光学遥感的潮沟精细提取方法

潮沟系统是粉砂淤泥质潮滩中最为活跃的地貌单元,受潮汐周期性冲刷、人类活动和海平面上升等因素的影响,大范围潮沟精细监测具有挑战性。本文提出了一种基于高分三号全极化合成孔径雷达(SAR)与PlanetScope多光谱遥感影像的潮沟探测和提取方法。通过融合光谱、指数、极化、纹理等特征,构建最优特征集,结合最大似然法、支持向量机及随机森林算法开展协同分类,获得了黄河口3 m分辨率的潮沟精细分布信息。结果表明,该方法的总体精度达到99%,F1值为0.98,提取结果优于单一数据源。本文方法有望为河口海岸带潮沟制图提供一种经济、有效的选择,有助于定量描述潮沟形态演变、稳定性及驱动因素。

光波导输入的双通道同步定标虚拟成像相位阵列光谱仪

激光频率梳定标的虚拟成像相位阵列光谱仪有潜力同时实现超高光谱分辨率与波长定标精度,对于天文光谱探测有重要意义。然而,现有的研究工作中,虚拟成像相位阵列光谱仪采用常规的双通道光纤输入模式,在受环境干扰时,两个通道的同步性较差,相对漂移严重,从而降低了光谱仪的重复性定标精度。为解决该问题,提出一种以光波导作为输入器件的双通道虚拟成像相位阵列光谱仪,利用波导通道之间相对稳定的特性,实现更高水平的同步定标精度。通过对不同环境下光谱仪的稳定性进行研究得出,在相似的测试条件下,波导输入的虚拟成像相位阵列光谱仪的重复性定标精度明显优于光纤输入时的定标表现。

A Learning-Based Channel Model for Synergetic Transmission Technology

It is extensively approved that Channel State Information(CSI) plays an important role for synergetic transmission and interference management. However, pilot overhead to obtain CSI with enough precision is a significant issue for wireless communication networks with massive antennas and ultra-dense cell. This paper proposes a learning- based channel model, which can estimate, refine, and manage CSI for a synergetic transmission system. It decomposes the channel impulse response into multiple paths, and uses a learning-based algorithm to estimate paths' parameters without notable degradation caused by sparse pilots. Both indoor measurement and outdoor measurement are conducted to verify the feasibility of the proposed channel model preliminarily.

Spectral Efficiency Analysis for Downlink Massive MIMO Systems with MRT Precoding

As an important part of the channel fading, large scale fading should be considered in downlink massive multiple-input multipleoutput(MIMO) systems. This paper investigates the downlink massive MIMO system over a large scale fading channel, when the base station takes advantage of maximum-radio transmission(MRT) precoding. The cases when the base station has perfect and imperfect channel state information(CSI) are considered respectively. Specially, through the analysis of single user's ergodic achievable rate, some novel and approximate expressions for the spectral efficiency(SE) are derived. Based on the approximate SE, the effect of the channel estimation error is analyzed intuitively. In addition, the average SE of all the users with different large-scale fading parameters is carefully investigated. Simulations validate that all the theoretical results coincide with numerical results and the large scale fading factors have little influence on SE reduction resulted from channel estimation.

基于多层特征融合的多光谱行人检测方法

针对行人检测模型在全天候场景下特征提取能力差、检测准确率低的问题,提出一种基于多层特征融合的多光谱行人检测方法。设计一种基于通道注意力机制的可见光与红外特征的融合方式,提升模型的特征融合效率;采用多层融合特征设计特征金字塔网络,提升模型的特征提取能力;引入自适应特征融合机制对检测层进行特征图尺度调整,降低尺度冲突对模型性能的影响。在KAIST数据集上进行实验,其结果表明,模型的检测性能有一定提升。

Spectral Characteristics of Wind-Waves in Laboratory

Spectral characteristics of wind-generated waves in labortaory are presented on the basis of a systematic measurement in a large-scale wind-wave channel and compared with those in the field. A marked characteristics of the measured spetra is the existence of secondary spectrum-peak. The dependence of spectral peak-frequency, peak-value and zeroth-order moment on wind speed and fetch are presented and found roughly similar to those in the field represented by the JONSWAP spectrum, regardless of the differences in coefficient. The spectral slope beta at high-frequencies are found somewhat greater than those of field wind-waves in both cases of deep and shallow waters. Except for the low-frequency part, the spectral forms measured in different wind conditions are similar and fit for the JONSWAP spectrum with gamma = 6 and beta = 5.5. Some relevant problems are discussed.

PERFORMANCES OF ADAPTIVE MODULATION AND CODING IN SATELLITE MOBILE CHANNEL

Given that satellite mobile channel is a time-varying channel,Adaptive Modulation and Coding(AMC) was proposed to provide robust and spectrally efficient transmission over satellite mobile channel.Three different kinds of channel environment were considered in this paper:the urban area,the rural area,and the open space.Four combinations of modulation and coding were designed to meet reliable communication on time-varying channel,and spectral efficiency and system throughput of these three kinds of channel environment were simulated.Based on the simulation results,this paper analysed the results and compared the performances of AMC with non-AMC system in these three kinds of channel environment.At last,we come to the conclusions:a system with AMC can achieve higher spectral efficiency and better system throughput;and the spectral efficiency and system throughput of AMC system will be higher on better satellite mobile channel.

Application of Seismic Attributes for Delineation of Channel Geometries and Analysis of Various Aspects in Terms of Lithological and Structural Perspectives of Lower Goru Formation, Pakistan

Miano area is one of the distinct major hydrocarbon producing fields of the Lower Indus basin. Lower Goru is the reservoir strait in this area. The aim of the present study is to exploit the channels reservoirs and other stratigraphic features in such a terrain where there is always a challenge for the geoscientist to search and exploit the channeled reservoirs. To resolve this issue we have utilized attribute analysis on high resolution 3D-seismic data for the detailed comparative studies for the channels. There are many astonishing features that are identified in the current study, which could not otherwise be easily enhanced with the help of 2D Seismic Data. Seismic attributes such as coherency, frequency (are also appropriate for lithological discrimination), which are sensitive to the channel edges are applied for the channel delineation and their geometrical analysis. Spectral decomposition techniques are also applied for the delineation of channels and to appropriately select the best band for channels identification. Three types of channel geometries are recognized: 1) highly sinuous channel;2) narrow-broad meandering belts;3) moderate to high sinuous channel. NW-SE, N-W trending faults can be helpful to compartmentalize the reservoir. Instantaneous and dominant frequency are more beneficial for further field development based on Gamma Ray logs from nearby drilled wells and dimensional perspectives analysis of the channel reservoir.

改进型CenterNet的高铁无砟轨道板表面裂缝检测算法

针对传统的高铁无砟轨道板表面裂缝检测方法存在检测精度低、速度慢的问题,提出一种改进型CenterNet的高铁无砟轨道板表面裂缝检测算法。该算法在编解码网络之间加入空洞空间金字塔池化模块(ASPP),以此扩大特征图的感受野,充分提取不同尺度的上下文信息;然后在特征提取网络中加入多光谱通道注意力模块(MCA),使网络可以更好学习每个通道的权重,捕获图像丰富的输入特征信息;最后使用α-IoU损失函数来提高边界框预测的准确度。实验结果表明,本算法平均检测精度(mAP)达到84.12%,相比传统算法平均检测精度提升了3.37%,对于轨道板表面裂缝具有较好的检测效果。

线性调频激光自混合干涉双通道微位移测量方法研究

为了满足工程中一些特殊二维位移测量需求或提高位移测量效率,建立了结构简单紧凑的激光自混合干涉双通道位移测量系统。首先,基于三镜法-珀腔模型给出自混合干涉系统的数学方程。其次,在弱反馈条件下施加线性电流调制,依据自混合信号频率和外部物体距离的线性关系,当两个物体到激光器距离不同时,频域会呈现两个独立的谱峰,分别对其进行相位解算,从而实现自混合双通道位移测量。然后,数值模拟生成了双通道激光自混合信号,根据全相位频谱分析技术对自混合信号两个谱峰的相位进行估算,重构了两个物体位移曲线,给出了仿真验证。最后,搭建了实验系统,进行了自混合干涉双通道位移测量实验,并给出实验测量结果。实验结果表明,该系统可以完全区分两个运动物体,位移测量相对误差优于8.42%。线性调频激光自混合干涉可以实现任意运动规律的双通道位移测量,通过继续分光其测量通道数仍可进一步增加。

基于频谱叠加技术的无线侧信道分析方法

在传统的侧信道分析过程中,研究者们在采集信号时需要和目标设备保持极短距离,这使得侧信道分析方法在很多场景中失去了现实意义.于2018年被提出的无线侧信道分析方法则能在数米的距离采集目标设备产生的侧信道信号,成功突破了侧信道分析方法在距离方面的限制.然而,无线侧信道分析方法易受到噪声的干扰.在高噪声环境中,该方法无法采集到良好的信号曲线,进而导致分析失败.为了解决这一困境,本文提出了基于频谱叠加技术的无线侧信道分析方法.该方法叠加多个频段的无线侧信道信号,能够有效消除信号中的高斯噪声,并提升信号信噪比与分析效率.本文建立了相应的信号传播模型,利用该方法在噪声环境中对加密设备进行分析.实验结果表明,频谱叠加技术能够有效地提升信号信噪比,对于侧信道分析效率的提升效果高达31.4%.因此,基于频谱叠加技术的无线侧信道分析方法优于原始的无线侧信道分析方法,成功降低了外界噪声的干扰.

自由空间光通信中的光广义正交空间调制技术

为了提高现有光空间调制系统的频谱效率,提出了一种在同相和正交域同时激活多根发射光学天线分别传输调制信号的实部与虚部的光广义正交空间调制(OGQSM)技术。该技术将调制符号扩展到同相和正交域,使得系统在空间域可以携带更多的比特,从而提高了系统频谱效率和误码性能。采用适用于所有湍流状态的Malaga湍流信道,分析了OGQSM系统在大气湍流、指向误差和路径损耗联合作用下的系统误码性能。蒙特卡罗仿真结果表明:与现有的光空间调制技术相比,所提出的OGQSM技术能够获得更好的频谱效率和误码性能,且接收光学天线数量与系统误码性能存在正相关关系,指向误差和大气湍流强度与系统误码性能存在负相关关系。

三维探地雷达通道不平衡抑制方法研究

近年来,我国城市道路塌陷事故频繁发生,造成了严重的人员伤亡和巨大的经济损失,这些塌陷大多是由地下空洞、脱空、疏松等隐患演化形成的,因此,定期对道路进行地下塌陷隐患检测并及时修复具有重要意义。目前,三维探地雷达已成为城市道路塌陷隐患检测公认最有效的地球物理方法,具有无损性、采集速度快、分辨率高和图像直观易解译等优点。然而,受限于三维探地雷达所采用阵列式天线的制作工艺、电子器件性能差异以及通道间串扰等影响,三维探地雷达数据存在通道不平衡问题,降低了合成三维图像的成像质量,进而影响塌陷隐患的识别和评估。本文针对三维探地雷达中存在的通道不平衡问题,提出了一种改进的谱白化方法,能有效减轻通道间的不平衡现象,提高三维探地雷达的雷达图像质量。

毫米波大规模MIMO系统联合稀疏混合预编码优化算法

为提高毫米波大规模多输入多输出(multiple input multiple output,MIMO)系统频谱效率及误码率性能,提出了一种联合稀疏混合预编码优化算法.首先根据毫米波信道的角度稀疏特性重构波瓣信道,分别对每个波瓣子信道进行混合预编码优化设计;然后针对混合预编码设计中含有非凸约束的多元稀疏信号重建问题,利用数字预编码矩阵的隐含稀疏结构,设计波瓣内每个数据流模拟预编码的自有支撑集和共有支撑集;最后根据模拟预编码矩阵与数字预编码矩阵之间的相关性,联合优化设计模拟预编码和数字预编码矩阵.进行了计算复杂度分析以及仿真分析.结果表明:与OMP算法相比,文中算法计算复杂度降低了91%或93%;相较于SLD算法,文中算法的频谱效率性能提高、误码率性能更好.