| 시간 | 제목 | 좌장/연사 |
| 09:00 ~ 10:15 | Stochastic geometry and wireless networks - a survey | 발표자: Francois Baccelli (INRIA) 좌장: 이남윤 (고려대) |
| 10:30 ~ 11:45 | Stochastic geometry and wireless networks - new directions | |
| 요약 : Stochastic Geometry is now commonly used for analyzing spectrum sharing in large wireless networks. In this approach, network elements, such as users and base stations, are represented as point processes in the Euclidean plane, and interference fields as spatial shot-noise processes. The analytical machinery of stochastic geometry and basic formulas of information theory can then be combined to predict important spatial statistics of such networks. The first part of the talk will cover the first steps of this approach which are based on the notions of SINR cells and SINR graphs associated with a Poisson point process. It will also cover the derivation of the distribution of the Shannon rates obtained by users in the Poisson-Voronoi model, which is the simplest mathematical abstraction for large cellular networks. In this model, base stations are represented by a Poisson point process and users connect to the closest base station. The most important recent advances on these basic models will also be surveyed: scaling laws, local statistics, network information theoretic extensions. The second part will show that one can introduce various types of dynamics in this stochastic geometry framework: user and base station motion, join and leave dynamics, queuing dynamics, and local adaptive dynamics. The results on this class of questions will be exemplified on the join and leave case through a discussion of the wireless birth-and-death model. In this model, users arrive according to a Poisson rain process on the Euclidean plane and leave with a stochastic intensity proportional to their instantaneous Shannon rate. The main results on the other types of dynamics will also be surveyed. | ||
| 11:45 ~ 13:30 | Lunch | |
| 13:30 ~ 14:00 | Satellite Downlink Coverage Analysis: From Binomial to Poisson Point Pro- cesses | 발표자: 이남윤 (고려대) 좌장: 박정훈 (연세대) |
| 요약 : Satellite networks are promising to provide ubiquitous and high-capacity global wireless connectivity. Traditionally, satellite networks are modeled by placing satellites on a grid of multiple circular orbit geometries. Such a network model, however, requires intricate system-level simulations to evaluate coverage performance, and analytical understanding of the satellite network is limited. Continuing the success of stochastic geometry in a tractable analysis for terrestrial networks, in this talk, we develop novel models that are tractable for the coverage analysis of satellite networks using stochastic geometry. By modeling the locations of satellites and users using Poisson point processes on the surfaces of concentric spheres, we characterize analytical expressions for the coverage probability of a typical downlink user as a function of relevant parameters, including path-loss exponent, satellite height, density, and Nakagami fading parameter. Then, we also derive a tight lower bound of the coverage probability in tractable expression while keeping full generality. Leveraging the derived expression, we identify the optimal density of satellites in terms of the height and the pathloss exponent. Our key finding is that the optimal average number of satellites decreases logarithmically with the satellite height to maximize the coverage performance. Simulation results verify the exactness of the derived expressions. We also present some recent results on the coverage probability gain for cooperative beamforming of satellites. | ||
| 14:00 ~ 14:30 | Coverage Analysis for Downlink Satellite Networks: Effect of Shadowing | 발표자: 최진석 (KAIST) 좌장: 박정훈 (연세대) |
| 요약 : Satellite communications have been promising to guarantee global coverage with high capacity. In this seminar, we share the analysis results of the coverage performance of satellite networks with a distance-dependent line-of-sight (LOS) and non-LOS (NLOS) channel propagation probability to incorporate shadowing effect. Extending the stochastic geometry-based network analysis for terrestrial networks, we model the satellite network and users as a Poisson point process and derive an theoretical coverage probability expression to provide analytical understanding of the satellite network. Simulation results verify the exactness of the derived expression. The derived expression includes network parameters for satellite density and altitude, channel fading, pathloss, and the LOS probability, and provides insights on satellite networks. Our key finding is that NLOS channel propagation benefits the coverage performance by reducing the interference from non-associated satellites, and the higher NLOS probability is desirable to improve the coverage performance as the network becomes denser. | ||
| 15:30 ~ 16:00 | Coverage Analysis of LEO Satellite Downlink Networks: Orbital Geometry Dependent Approach | 발표자: 이준세 (성신여대) 좌장: 박정훈 (연세대) |
| 요약 : The low-earth-orbit (LEO) satellite network has emerged as a promising technology for providing global coverage with high-data rates. However, the performance analysis of such a network is very challenging because it is highly dependent on several orbit geometry parameters, including satellite altitude and inclination angle. This talk introduces a novel systematic analysis framework for the LEO satellite network, which highlights the importance of orbit geometric parameters. We propose a model where satellite locations are distributed as a onedimensional Poisson point process on a circular orbit. Then, we derive the distribution of the nearest distance between a satellite and a fixed user’s location on Earth as a function of the orbitgeometry parameters. By leveraging this distribution, we characterize the coverage probability of a single-orbit LEO network and then extend our coverage analysis to multi-orbit networks. We demonstrate the benefits of harnessing multi-orbit satellite networks in terms of coverage probability. | ||
| 16:00 ~ 16:30 | 휴식 | |
| 16:30 ~ 17:00 | An Analytical Framework for Downlink LEO Satellite Communications based on Cox Point Processes | 발표자: 최창식 (홍익대) 좌장: 최진석 (KAIST) |
| 요약 : This work develops an analytical framework for downlink low earth orbit (LEO) satellite communications, leveraging tools from stochastic geometry. We propose a tractable approach to the analysis of such satellite communication systems accounting for the fact that satellites are located on circular orbits. We accurately characterize this geometric property of such LEO satellite constellations by developing a Cox point process model that jointly produces orbits and satellites on these orbits. Our work differs from existing studies that have assumed satellites’ locations as completely random binomial point processes. For this Cox model, we derive the outage probability of the proposed network and the distribution of the signal-to-interferenceplus-noise ratio (SINR) of an arbitrarily located user in the network. By determining various network performance metrics as functions of key network parameters, this work allows one to assess the statistical properties of downlink LEO satellite communications and thus can be used as a system-level design tool. | ||
| 17:10 ~ 17:40 | Connectivity from Earth and Space: Modeling and Coverage Analysis for Space-Terrestrial Integrated Networks with Stochastic Geometry | 발표자: 박정훈 (연세대) 좌장: 최진석 (KAIST) |
| 요약 : Recently, the integration of terrestrial and satellite networks (STIN), where frequency bands are shared to LEO satellite networks and terrestrial networks, is of great interest in not only expanding coverage region in remote rural areas where no terrestrial base station is deployed, but also providing potential load balancing in dense urban areas. Since the LEO satellites induce mutual interference to the terrestrial network’s user and vice versa in the STIN, the analysis on the STIN is much more complicated and challenging than a single LEO network or a single terrestrial network; yet the analysis is crucial to provide network design guidelines for the STIN. In this talk, we present novel modeling and analysis techniques for STIN. Upon the presented network model, we investigate the rate coverage performance of the STIN and understand the STIN benefits more clearly. By doing so, we offer valuable insights into the potential of STIN for improving network coverage and load balancing. | ||
