Ray-tracing based analysis of channel characteristics and capacity improvement capabilities of spatial multiplexing and beamforming at 15 and 28 GHz

Sammanfattning: In the previous decades, mm-Wave frequency bands have largely been ignored for cellular communication because of the high free space attenuation as well as the cost and complexity of mm-Wave RF circuits. However, with the improvement of RF and antenna technologies over the last decade, it has become feasible to con- sider mm-Wave bands for cellular communications. With the large bandwidths available in the mm-Wave range, this could enable much higher data rates, and could also alleviate the problem of limited frequency resources. Many bands in the frequency range above 6 GHz and into the mm-Wave range of 30-300 GHz are therefore interesting candidates for future 5G cellular systems. The utilisation of Multiple-input Multiple-output (MIMO) is almost a necessity at these frequencies. This makes it possible to implement different antenna processing techniques in order to improve coverage and system capacity. In the master thesis, the behaviour of 15 and 28 GHz channels in indoor and outdoor scenarios are simulated by us- ing a ray-tracing algorithm. By comparing indoor channel measurements with the ray-tracing results, it is possible to verify the performance and identify the limitations of the ray-tracing algorithm. In the outdoor scenario, channel characteristics such as received power and RMS delay spreads are analysed based on ray-tracing simulations. A metric called Channel Multiplexing Richness (CMR) is defined to indicate the scattering richness of a location. This thesis also investigates the capacity improving capabilities of beamforming (such as directional beamforming and dominant eigenmode transmission), spatial multiplexing and a hybrid technique which combines beamforming and spatial multiplexing. The link capacity for these different techniques is calculated based on ray tracing results for the single-user case, assuming full channel state information. The simulation results are also compared to different environment models, so that the influences of the material permittivity and the level of geometric details are investigated.