New Fusion Approach of Spatial and Channel Attention for Semantic Segmentation of Very High Spatial Resolution Remote Sensing Images

The semantic segmentation of very high spatial resolution remote sensing images is difficult due to the complexity of interpreting the interactions between the objects in the scene. Indeed, effective segmentation requires considering spatial local context and long-term dependencies. To address this problem, the proposed approach is inspired by the MAC-UNet network which is an extension of U-Net, densely connected combined with channel attention. The advantages of this solution are as follows: 4) The new model introduces a new attention called propagate attention to build an attention-based encoder. 2) The fusion of multi-scale information is achieved by a weighted linear combination of the attentions whose coefficients are learned during the training phase. 3) Introducing in the decoder, the Spatial-Channel-Global-Local block which is an attention layer that uniquely combines channel attention and spatial attention locally and globally. The performances of the model are evaluated on 2 datasets WHDLD and DLRSD and show results of mean intersection over union (mIoU) index in progress between 1.54% and 10.47% for DLRSD and between 1.04% and 4.37% for WHDLD compared with the most efficient algorithms with attention mechanisms like MAU-Net and transformers like TMNet.

Single-image night haze removal based on color channel transfer and estimation of spatial variation in atmospheric light

The visible-light imaging system used in military equipment is often subjected to severe weather conditions, such as fog, haze, and smoke, under complex lighting conditions at night that significantly degrade the acquired images. Currently available image defogging methods are mostly suitable for environments with natural light in the daytime, but the clarity of images captured under complex lighting conditions and spatial changes in the presence of fog at night is not satisfactory. This study proposes an algorithm to remove night fog from single images based on an analysis of the statistical characteristics of images in scenes involving night fog. Color channel transfer is designed to compensate for the high attenuation channel of foggy images acquired at night. The distribution of transmittance is estimated by the deep convolutional network DehazeNet, and the spatial variation of atmospheric light is estimated in a point-by-point manner according to the maximum reflection prior to recover the clear image. The results of experiments show that the proposed method can compensate for the high attenuation channel of foggy images at night, remove the effect of glow from a multi-color and non-uniform ambient source of light, and improve the adaptability and visual effect of the removal of night fog from images compared with the conventional method.

Statistical Channel Modeling for Indoor VLC Communications Based on Channel Measurements

Visible light communication(VLC)has attracted much attention in the research of sixthgeneration(6G)systems.Furthermore,channel modeling is the foundation for designing efficient and robust VLC systems.In this paper,we present extensive VLC channel measurement campaigns in indoor environments,i.e.,an office and a corridor.Based on the measured data,the large-scale fading characteristics and multipath-related characteristics,including omnidirectional optical path loss(OPL),K-factor,power angular spectrum(PAS),angle spread(AS),and clustering characteristics,are analyzed and modeled through a statistical method.Based on the extracted statistics of the above-mentioned channel characteristics,we propose a statistical spatial channel model(SSCM)capable of modeling multipath in the spatial domain.Furthermore,the simulated statistics of the proposed model are compared with the measured statistics.For instance,in the office,the simulated path loss exponent(PLE)and the measured PLE are 1.96and 1.97,respectively.And,the simulated medians of AS and measured medians of AS are 25.94°and 24.84°,respectively.Generally,the fact that the simulated results fit well with measured results has demonstrated the accuracy of our SSCM.

Spatial Correlation Characteristics Analysis of Multi-Beam Channels of Mobile Satellite System

Due to the influence of scatterers around the receiving antenna, the multipath signal in satellite mobile communication systems is correlated with each other which would influence the system performance. There is no systematic standard on the channel modelling of the wideband satellite channel at present, so the study of the modelling of the wideband satellite channel is of great importance. In this paper, firstly we created a multi-beam model which can figure out the antenna gain of the nth component beam. Secondly, we combined the characteristics of multi-beam satellite channel and the distribution of the scatterers, and set up a three-dimension random channel model. This model is more realistic for satellite communication system since it considers the height of scatterers. According to the channel models, we had the formula of spatial correlation coefficient. We used the formula to calculate the relationship between spatial correlation coefficient and the interval of antennas. The result shows that the spatial correlation exists and cannot be ignored while modeling for mobile satellite system.

Performance Improvement in Estimation of Spatially Correlated Rician Fading MIMO Channels Using a New LMMSE Estimator

In most of the previous researches on the multiple-input multiple-output (MIMO) channel estimation, the fading model has been assumed to be Rayleigh distributed. However, the Rician fading model is suitable for microcellular mobile systems or line of sight mode of WiMAX. In this paper, the training based channel es-timation (TBCE) scheme in the spatially correlated Rician flat fading MIMO channels is investigated. First, the least squares (LS) channel estimator is probed. Simulation results show that the Rice factor has no effect on the performance of this estimator. Then, a new linear minimum mean square error (LMMSE) technique, appropriate for Rician fading channels, is proposed. The optimal choice of training sequences with mean square error (MSE) criteria is investigated for these estimators. Analytical and numerical results show that the performance of proposed estimator in the Rician channel model compared with Rayleigh one is much better. It is illustrated that when the channel Rice factor and/or the correlation coefficient increase, the per-formance of the proposed estimator significantly improves.

STABILITY ANALYSIS IN SPATIAL MODE FOR CHANNEL FLOW OF FIBER SUSPENSIONS

Different from previous temporal evolution assumption, the spatially growing mode was employed to analyze the linear stability for the channel flow of fiber suspensions. The stability equation applicable to fiber suspensions was established and solutions for a wide range of Reynolds number and angular frequency were given numerically . The results show that, the flow instability is governed by a parameter H which represents a ratio between the axial stretching resistance of fiber and the inertial force of the fluid. An increase of H leads to a raise of the critical Reynolds number, a decrease of corresponding wave number, a slowdown of the decreasing of phase velocity , a growth of the spatial attenuation rate and a diminishment of the peak value of disturbance velocity. Although the unstable region is reduced on the whole, long wave disturbances are susceptible to fibers.

Single foggy image restoration based on spatial correlation analysis of dark channel prior

Focusing on the degradation of foggy images, a restora- tion approach from a single image based on spatial correlation of dark channel prior is proposed. Firstly, the transmission of each pixel is estimated by the spatial correlation of dark channel prior. Secondly, a degradation model is utilized to restore the foggy image. Thirdly, the final recovered image, with enhanced contrast, is obtained by performing a post-processing technique based on just-noticeable difference. Experimental results demonstrate that the information of a foggy image can be recovered perfectly by the proposed method, even in the case of the abrupt depth changing scene.

Investigations into the Effect of Spatial Correlation on Channel Estimation and Capacity of Multiple Input Multiple Output System

The paper reports on investigations into the effect of spatial correlation on channel estimation and capacity of a multiple input multiple output (MIMO) wireless communication system. Least square (LS), scaled least square (SLS) and minimum mean square error (MMSE) methods are considered for estimating channel properties of a MIMO system using training sequences. The undertaken mathematical analysis reveals that the accuracy of the scaled least square (SLS) and minimum mean square error (MMSE) channel estimation methods are determined by the sum of eigenvalues of the channel correlation matrix. It is shown that for a fixed transmitted power to noise ratio (TPNR) assumed in the training mode, a higher spatial correlation has a positive effect on the performance of SLS and MMSE estimation methods. The effect of accuracy of the estimated Channel State Information (CSI) on MIMO system capacity is illustrated by computer simulations for an uplink case in which only the mobile station (MS) transmitter is surrounded by scattering objects.

Spatial Channel Sounding Based on Bistatic Synthetic Aperture Radar Principles

In this paper, we propose a simplified spatial channel sounding method by utilizing bistatic synthetic aperture radar(BiSAR) principles. Despite the different deployment geometries compared with a conventional BiSAR system, the feasibility of the approach is established by 1) the proposed method achieves a better spatial resolution than conventional directional channel sounders and 2) reconstruction algorithms based on time-domain backprojection in conjunction with a digital elevation model provide a good imaging performance and are suitable for reconstructing the spatial distribution of scatterers. Simulations of a high-speed rail(HSR) scenario demonstrate that the estimated power delay profiles(PDPs) and power angle profiles(PAPs) are close to the actual values.

Linear Precoder Design for Correlated Rician Fading Channel Under Imperfect CSI

In this paper,we have modeled a linear precoder for indoor multiuser multiple input multiple output(MU-MIMO)system with imperfect channel state information(CSI)at transmitter.The Rician channel is presumed to be mutually coupled and spatially,temporarily correlated.The imperfection with CSI is primarily due to the channel estimation error at receiver and feedback delay amidst the receiver and transmitter in CSI transmission.Along with,the insufficient spacing between the antenna at transmitter and receiver persuades mutual coupling(MC)among the array elements.In addition,the MIMO channel is presumed to be jointly correlated(Weichselberger correlation model).When we look back on the existing precoder design,it considered spatial correlation alone disregarding joint correlation of antenna array elements.With all above assumption,we have designed a linear precoder which minimizes mean squared error(MSE)subjected to total transmit power constraint for MUMIMO system.The simulation results proven that proposed precoder shows substantial enhancement in bit error rate(BER)performance in comparison with the existing technique.The mathematical analysis corroborates the simulation results.

Measurement.Based Spatial-Consistent Channel Modeling Involving Clusters of Scatterers

In this paper,the conventional method of establishing spatial channel models(SCMs)based on measurements is extended by including clusters-of-scatterers(CoSs)that exist along propagation paths.The channel models resulted utilizing this new method are applicable for generating channel realizations of reasonable spatial consistency,which is required for designing techniques and systems of the fifth generation wireless communications.The scatterers’locations are estimated from channel measurement data obtained using large-scale antenna arrays through the Space-Alternating Generalized Expectation-Maximization(SAGE)algorithm derived under a spherical wavefront assumption.The stochastic properties of CoSs extracted from real measurement data in an indoor environment are presented.

MIMO-OFDM Scheme Using ApFFT over 3GPP SCM Channels

The all-phase fast Fourier transform (apFFT) is proposed as a digital demodulation algorithm in place of the fast Fourier transform (FFT) for orthogonal frequency division multiplexing (OFDM) based multiple-input multiple-output (MIMO) communication systems. The amplitude spectrum of apFFT-demodulated symbols is the square of that of the FFT, which helps reduce the Gaussian noise to a great extent. Moreover, the phases of apFFT symbols are not affected by the frequency shift between the transmitter and receiver oscillators. These properties particularly appeal to MIMO systems over frequency-selective fading channels. The proposed MIMO-OFDM system employing the apFFT is validated using the spatial channel model (SCM) proposed by the third generation partnership project (3GPP). The simulation results demonstrate that the performance of the proposed system after compensating for the rate loss due to zero bits inserted in the space-frequency OFDM (SF-OFDM) coding scheme, still considerably outperforms the conventional system over 3GPP SCM channels, especially under poor channel conditions.

Elimination of Spatial Side-Channel Information for Compact Quantum Key Distribution Senders

For a compact quantum key distribution (QKD) sender for the polarization encoding BB84 protocol, an eavesdropper could take a side-channel attack by measuring the spatial information of photons to infer their polarizations. The possibility of this attack can be reduced by introducing an aperture in the QKD sender, however, the effect of the aperture on the QKD security lacks of quantitative analysis. In this paper, we analyze the mutual information between the actual keys encoded at this QKD sender and the inferred keys at the eavesdropper (Eve), demonstrating the effect of the aperture to eliminate the spatial side-channel information quantitatively. It shows that Eve’s potential on eavesdropping spatial side-channel information is totally dependent on the optical design of the QKD sender, including the source arrangement and the aperture. The height of compact QKD senders with integrated light-emitting diode (LED) arrays could be controlled under several millimeters, showing great potential on applications in portable equipment.

A novel 3-D spatial temporal channel model for high mobility HAPS-based communications scenario

A high altitude platform station (HAPS) based communications scenario for providing Internet accessand broadband multimedia services to the passengers on board of a high speed train (traveling up to300km/h) is proposed. Regarding the addressed scenario, when the propagation link between HAPS andtrain is blocked by obstacles, a three-dimensional (3-D) geometrical single cylinder spatial-temporalchannel model is presented, in which closed form, mathematically tractable space-time correlation functionsare obtained. It shows that the correlation functions determined by the 3-D model are of significantdifference with those of the conventional 2-D model. Based on the analysis model, the paper derives a realizedsimulation model using sum-of-sinusoids approach, and applies method of equal areas (MEA) andmodified method of equal areas (MMEA) to determine the model parameters. The fitting performance ofthe simulation model with the analysis one is evaluated by two means-square error (MSE) performancecriteria. Finally, numerical simulation results verify the mathematical analysis conclusion, when N≥21,simulation model has an excellent fitness with the analysis one.

MmWave extra-large-scale MIMO based active user detection and channel estimation for high-speed railway communications

The current High-Speed Railway(HSR)communications increasingly fail to satisfy the massive access services of numerous user equipment brought by the increasing number of people traveling by HSRs.To this end,this paper investigates millimeter-Wave(mmWave)extra-large scale(XL)-MIMO-based massive Internet-of-Things(loT)access in near-field HSR communications,and proposes a block simultaneous orthogonal matching pursuit(B-SOMP)-based Active User Detection(AUD)and Channel Estimation(CE)scheme by exploiting the spatial block sparsity of the XLMIMO-based massive access channels.Specifically,we first model the uplink mmWave XL-MIMO channels,which exhibit the near-field propagation characteristics of electromagnetic signals and the spatial non-stationarity of mmWave XL-MIMO arrays.By exploiting the spatial block sparsity and common frequency-domain sparsity pattern of massive access channels,the joint AUD and CE problem can be then formulated as a Multiple Measurement Vectors Compressive Sensing(MIMV-CS)problem.Based on the designed sensing matrix,a B-SOMP algorithm is proposed to achieve joint AUD and CE.Finally,simulation results show that the proposed solution can obtain a better AUD and CE performance than the conventional CS-based scheme for massive IoT access in near-field HSR communications.

复杂战场环境下改进YOLOv5军事目标识别算法研究

复杂战场环境下军事目标识别技术是提升战场情报获取能力的基础和关键。针对当前军事目标识别技术在复杂战场环境下漏检误检率高、实时性差等问题,提出一种基于改进YOLOv5模型的PB-YOLO军事目标识别算法。将改进的目标识别算法对于陆战场军事单元的识别锚框进行重新聚类,以提升模型对于目标大小适应度,加速模型收敛;采用通道-空间并行注意力机制,增加模型对复杂战场环境下目标特征信息与位置信息关注度;在特征融合网络部分使用BiFPN以提升模型对于特征的融合能力与速度;采用Alpha_IoU损失函数加速模型收敛,解决当真实框与预测框重合时IoU计算退化问题。实验结果表明,在自建军事目标数据集下,改进算法与主流目标识别算法相比,在保证模型空间复杂度的同时,mAP值达到了90.17%。消融实验对比结果表明,改进后网络较原模型精度提升11.57%,具有较好的识别性能,能够为战场情报获取提供有效的技术支撑。

基于注意力机制及多分支特征融合的实时语义分割算法

为了合理平衡语义分割中的精确度与实时性,基于快速卷积神经网络模型(Fast-SCNN)提出了一种基于注意力机制及多分支特征融合的实时语义分割算法模型。该算法模型首先通过注意力模块捕获空间特征之间的相互联系,增强空间细节信息;然后合理设计融合模块,最大化利用各分支信息,实现深层特征与浅层特征更好的融合;最后引入自适应特征增强注意力模块,捕获长距离像素间的相互依赖关系。实验结果表明,文中算法模型在Cityscapes数据集上获得了71.55%的分割精度,推理速度FPS达到97.6帧/s,模型参数量为1.39 M,验证了该算法所构成网络模型的有效性。

基于混合域残差注意力网络的滚动轴承智能故障诊断方法

机械设备正朝着大型化、精密化和自动化的方向发展,机械系统也因此变得越来越复杂。考虑到机械系统可能会发生无特征的灾难性故障,因此机械故障的自动检测是一个巨大的挑战。然而,现有的故障检测方法在对高度复杂的工业系统进行故障类型识别时,误诊率较高,无法给出准确的故障诊断结果。针对这一问题,本文以滚动轴承这一机械设备关键部件作为研究对象,提出一种基于混合域残差注意力网络的故障诊断方法,旨在结合深度卷积神经网络自动学习表示的优点,并配合通道注意力机制和空间注意力机制的关键特征提取能力,提高故障检测性能。实验结果表明,所提出的方法能够准确地检测轴承故障类型,在准确度指标方面优于其他方法。

融合双分支语义增强感知的遥感图像超分辨率重建算法

针对遥感图像中地物目标的特征信息模糊以及背景噪声影响导致遥感图像重建效果差的问题,提出一种融合双分支语义增强感知的遥感图像超分辨率重建算法。首先,设计了一种全局-局部空间注意力模块,该模块用于增强特征在空间全局-局部不同尺度下的语义表征能力,同时强化网络对有效特征组的分辨能力;其次,提出一种通道分组-聚合注意力模块,通过设计特征分组-聚合以及通道注意力模块,增强模型对地物目标特征的区分,强化对有效特征通道的关注能力。实验表明,所提算法在UC Merced数据集上,峰值信噪比在×2/×3/×4倍率下分别达到了34.397 dB、29.920 dB和28.128 dB,结构相似度在×2/×3/×4倍率下达到了0.931、0.834和0.791。在AID数据集上,峰值信噪比在×2/×3/×4倍率下分别达到了32.524 dB、29.317 dB和27.522 dB,结构相似度在×2/×3/×4倍率下达到了0.895、0.829和0.721。两个指标相较于等主流算法均有所提升,重建后图像的边缘与区域细节效果更优,有效克服了地物目标的特征信息模糊及背景噪声影响导致遥感图像重建效果差的问题。

基于可靠性分析的避震疏散通道通达性评价

城市避震疏散通道的震后通达性直接关系到能否在第一时间安全转移受灾群众及顺利开展救灾工作。以济南市杆石桥街道为研究对象,考虑到震中沿街建筑的倒塌影响,借助空间句法进行道路系统的抗震可靠性和通达性评价,并提出优化建议。经分析,杆石桥街道建筑密度较高,高层建筑多布局于道路交叉口,均对街区道路的避震疏散功能有一定威胁,且老城区本身避震疏散网络规划不足、存在较多救灾“死角”。随着城市的发展和更新,大多中心城区的规划年代久远,原有避震疏散网络已不能满足需求,近年来自然灾害频发,中心城区的避震疏散通道亟待进行整体评价及优化。