Device-to-device communication and wearable networks harnessing spatial proximity

Spatially proximal devices wanting to exchange information are expected to become more prevalent in wireless networks, rendering the option for direct device-to-device (D2D) communication increasingly important. On the one hand, within networks where communication via infrastructure has been the con...

Descripción completa

Detalles Bibliográficos
Autor: George, Geordie
Tipo de recurso: tesis doctoral
Estado:Versión publicada
Fecha de publicación:2017
País:España
Institución:CBUC, CESCA
Repositorio:TDR. Tesis Doctorales en Red
OAI Identifier:oai:www.tdx.cat:10803/404986
Acceso en línea:http://hdl.handle.net/10803/404986
Access Level:acceso abierto
Palabra clave:Wireless networks
D2D communication
Ergodic spectral efficiency
Overlay
Underlay
Stochastic geometry
Poisson point process
Cellular networks
MIMO
Shadowing
Interference
SINR
Wearable networks
Millimeter wave
Indoor mmWave communication
Random shape theory
Directional beamforming
62
Descripción
Sumario:Spatially proximal devices wanting to exchange information are expected to become more prevalent in wireless networks, rendering the option for direct device-to-device (D2D) communication increasingly important. On the one hand, within networks where communication via infrastructure has been the convention, enabling such an option for short-range and single-hop communication between co-located devices might potentially bring about performance benefits on several accounts. On the other hand, in the realm of networks where direct interaction between devices has been an obvious option, there is a growing demand for supporting extreme-data-rate applications and much denser deployments of simultaneous transmissions. This dissertation explores these aspects by addressing two main problems: (i) analyzing the performance benefits of D2D communication integrated into cellular mobile networks, and (ii) investigating the feasibility of mmWave (millimeter wave) frequencies for personal networks of wearable (body-born) devices in enclosed settings. Under sufficient spatial locality in wireless traffic within cellular networks, the D2D mode of communication can be leveraged to employ a denser spectral reuse, thereby achieving very high area spectral efficiencies (bits/s/Hz per unit area). Enabling D2D entails a reshaping of the network topology comprising the sources of useful signal and harmful interference from the vantage of each receiver, which is a factor that delimits network performance fundamentally. Therefore, to gauge the performance gains of D2D and to identify the challenges thereof, it is essential to model D2D communication in a large multicellular setting, without missing key features of the ensuing interference environment. In this regard, we develop a robust analytical framework, utilizing tools from stochastic geometry. The dissertation propounds a novel approach to the application of stochastic geometry that is shown to improve the simplicity, accuracy, and generality of wireless network analysis. The performance evaluation conducted using the framework, while demonstrating the potential of D2D, also indicates the need for managing the interference surge. Prompted by this, and to illustrate the flexibility of the framework, we further extended it to incorporate interference protection schemes based on exclusion regions and the benefits thereof are assessed. The presence of multiple wearable networks—each comprising several on-body device-pairs worn by people—in proximity might result in an extreme density of simultaneous wireless transmissions. Such a scenario is expected to become commonplace in enclosed settings, e.g., commuter trains, subways, airplanes, airports or offices, and be further challenging due to an increasing demand for data-rate-intensive wireless applications in wearable technology. This combination of very-short-range communication, high-data-rate applications, and dense spectral reuse seems to render operation at mmWave frequencies a suitable candidate; add to that the possibility of accommodating antenna arrays within devices for directional beamforming. Hence, we investigate the feasibility of enclosed mmWave wearable networks, with a particular focus on appropriately modeling the impact of propagation mechanisms at these frequencies. In the propagation modeling, specular reflections off surfaces are explicitly accounted for, as they are expected to contribute useful signal power while, at the same time, intensify the interference. Recognizing the increased prominence of blocking by obstacles, body-blockages in the direct and reflected propagation paths are also modeled. The impact of these mechanisms on the spectral efficiency of the network is evaluated, aided by the application of stochastic geometry and random shape theory. Under relevant indoor settings, and in the plausible absence of strong direct signal, the reliability of surface reflections in providing useful signal power for efficient communication is investigated and the need for directional antennas is established.