The train-to-ground communication system is of vital importance to the safe and reliable operation of the vacuum tube high-speed train (vactrain). To better design the communication system, a full understanding of the channel characterization is essential and the accurate channel model needs to be investigated. In this paper, we propose a hybrid model of ray-tracing (RT) method and the propagation graph (PG) method based on directive scattering model and diffraction model. The line-of-sight (LoS) component, reflection, scattering, and diffraction components are considered. Based on the hybrid model, the expression of the channel transfer function (CTF) is derived and the power delay profile (PDP) and delay spread are obtained and analyzed. The simulation results show the wireless channel characteristics in vactrain scenarios and provide useful insights for future vactrain communication systems.

Low-Earth orbit (LEO) satellites have been considered essential to future air-space-ground integrated networks. Wireless channels significantly impact the performance of communication systems, especially in terms of large-scale fading characteristics. In this article, we propose a novel geometry-based stochastic channel model (GBSM) for LEO satellite-ground large-scale channels. Propagation probabilities of Line-of-Sight (LoS) links, ground specular links, and building specular links for suburban, urban, dense urban, and high-rise urban in different elevations are computed. The Fresnel zone is utilized to determine whether the signals can arrive at the receiver. The impact of the radio coverage and receiver height on propagation probabilities are considered for each scenario. Based on the derived propagation probabilities, the average path loss is computed. In the simulation section, the results of our model are validated by the Monte Carlo method, and the average path loss is compared with the standard model in 3GPP TR 38.811. The comparison results have good consistency with the standard model. Moreover, our model can be applied in multiple scenarios by adjusting the environment parameters compared with the standard model.

Hyperloop, a novel rail transportation technology, can run at a speed of more than 1000 km/h in the vacuum tube scenario. To guarantee the safe and reliable operation of the Hyperloop communication system, a new channel model considering the line-of-sight (LoS), reflection, and scattering components is proposed to investigate the channel characteristics of the Hyperloop scenarios. Based on the geometric relationship and physical propagation mechanism, the expression of channel gain is derived and the channel impulse response (CIR) is calculated. Then, the Doppler spectrum and K factor are analyzed. The simulation results show the wireless channel characteristics in the Hyperloop scenarios and provide useful insights for future Hyperloop communication systems.

As a novel communication pattern, non-reciprocal beams employ narrow-width beams to transmit signals and wide-width beams to receive, which can achieve fast beam alignment quickly. To fully understand the channel characterization, we investigate the mmWave propagation path loss (PL) over this special pattern. First, we present a beam-filtered power angular spectrum (PAS) simulation method based on the 3GPP. Specifically, the main lobe width-related beam gain is used to shape the generated PAS. On this basis, we consider two cases, i.e., perfect beam alignment and misalignment. For the first case, we propose a novel exponential decay PL model instead of the inversely proportional relationship. Furthermore, we verify our proposal based on the measured data of 28GHz and the results show that it yields a smaller fitting root mean square error (RMSE). Regarding the latter case, we find that the beam steering-based additional PL can be regarded as the superposition of a deterministic Gaussian function and a stochastic Gaussian noise. Our findings provide insightful references for the implementation of beamwidth-dependent nonreciprocal beam patterns.

Increasingly, cloud database vendors host large-scale geographically distributed clusters to provide cloud database services. When managing the clusters, we observe that it is challenging to simultaneously maximizing the resource allocation ratio and resource availability. This problem becomes more severe in modern cloud database clusters, where resource allocations occur more frequently and on a greater scale. To improve the resource allocation ratio without hurting resource availability, we introduce Eigen, a large-scale cloud-native cluster management system for large-scale databases on the cloud. Based on a resource flow model, we propose a hierarchical resource management system and three resource optimization algorithms that enable end-to-end resource optimization. Furthermore, we demonstrate the system optimization that promotes user experience by reducing scheduling latencies and improving scheduling throughput. Eigen has been launched in a large-scale public-cloud production environment for 30+ months and served more than 30+ regions (100+ available zones) globally. Based on the evaluation of real-world clusters and simulated experiments, Eigen can improve the allocation ratio by over 27% (from 60% to 87.0%) on average, while the ratio of delayed resource provisions is under 0.1%.

With the rapid growth of cloud computing, efficient management of multi-tenant databases has become a vital challenge for cloud service providers. It is particularly important for Alibaba, which hosts a distributed multi-tenant database supporting one of the world's largest e-commerce platforms. It serves tens of millions of sellers as tenants, and supports transactions from hundreds of millions of buyers. The inherent imbalance of shopping preferences from the buyers essentially generates a drastically skewed workload on the database, which could create unpredictable hotspots and consequently large throughput decline and latency increase. In this paper, we present the architecture and implementation of ESDB (ElasticSearch Database), a cloud-native document-oriented database which has been running on Alibaba Cloud for 5 years as the main transaction database behind Alibaba's e-commerce platform. ESDB provides strong full-text search and retrieval capability, and proposes dynamic secondary hashing as the solution for processing extremely skewed workloads. We evaluate ESDB with both simulated workloads and real-world workloads, and demonstrate that ESDB significantly enhances write throughput and reduces the completion time of writes without sacrificing query throughput.

In this paper, we performed a series of measurement campaigns on the wireless electromagnetic noise for two typical industrial welding scenarios. On this basis, we investigate the characterization of the impulsive noise mainly from two aspects, that is, frequency and time domains. To start with, a novel denoising method based on the dynamic threshold is proposed to identify the desirable impulse noise from the background noise. Then, in the frequency domain, we focus on the power distribution of impulsive noise at different frequency bands. Results exhibit a shadow effect with regard to different frequency bands and we characterize it by using a linear function with a Gaussian distribution. Besides, analyses on the power spectrum correlation for different polarization modes and scenarios are also provided. In the time domain, we performed a series of statistical analyses from aspects of pulse amplitude, duration, and elapse interval to characterize the impulsive noise. Furthermore, three empirical distributions are employed to depict the parameters' variation tendency, that is, Cauchy distribution for amplitude, Gamma distribution for pulse duration, and exponential distribution for pulse interval. Finally, a first-order two-states Markov method is proposed to model the industrial noise. Simulation results are proved to be consistent with the actual measured results in terms of the amplitude distribution.

The vacuum tube high-speed flying train (high-speed flying train) is a novel rail transportation technology. The maglev train can run with low mechanical friction, low air resistance and low noise mode at ultra-high-speed (over 1000 km/h) in all weather conditions inside the vacuum tube. In this paper, we describe the unique characteristics associated with the wireless communication of high-speed flying train, such as high Doppler frequency shift, metal waveguide effect and extreme frequent handoff. Therefore, the solution of leaky-wave system is selected to effectively suppress the Doppler effect in the vacuum tube. The simulation results show the even field distribution at the observation point far away from the leaky-wave system, and the uniform phrase distribution along the train motion direction, which can eliminate Doppler frequency shift. Since the simulation indicates that the conventional time-varying frequency-selective fading channel is converted into a stationary channel, we proposed an electromagnetic wave refractive lens system, which together with the leaky wave near-field convergence technology can directly achieve the even wireless signal coverage for passengers inside the train. In addition, in order to deal with the extremely frequent handoff, the solution of moving cell is adopted, which can be realized by the Centralized, cooperative, cloud Radio Access Network (C-RAN).

Channel denoising is of much importance for further channel characterization and fading analysis. In this paper, we proposed a novel method for the wireless channel impulse response (CIR) denoising based on neural network. Discriminating effective signals from the noise is considered as a binary classification problem, which can be resolved by some machine learning methods. The back propagation neural network (BPNN) model with the amplitude and angle information of CIR as input data is implemented to settle the classification problem. What's more, the Fβ-score, which combines both Precision and Recall together, is selected as the evaluation index to assess the classification performance. Compared with the performance of wavelet transform, the proposed method shows a better denoising result at both high and low signal-to-noise ratio (SNR) due to its full utilization of amplitude and angle information, while the BPNN with only amplitude as input data shows the worst result comparing with the other two methods.

A new approach based on conventional propagation graph channel modeling was proposed to haracterize the wireless channel in non-light of sight (NLOS) tunnel scenarios. The scattering points are regarded as several points sets, which are different from the propagation-graph theory, then the transfer probability among sets is introduced to adjust the channel impulse response (CIR) taps. The advantage of the proposed method is that wideband channel coefficients, CIR in delay, antennas' correlation coefficient, angle of arrival (AOA), angle of departure (AOD), channel capacity can be calculated analytically for these environments. The validation of the proposed method is performed by the reasonable distribution of the CIR taps, AOD and AOA. Finally some works are done to investigate the variation of tunnel channel coefficients when tunnel bending angle varies, and channel matrix degradation is adopted to explain it.