Simulations of the Capacity and Coverage for a Multi-User Distributed MIMO Network

Abstract

An ever growing increase of connected devices and data rates around the world needs to be met by new communication technologies. Using an antenna array with multiple antennas, a so-called MIMO system, has proven to be a successful way to boost the capacity of wireless communication systems. By spatially separating the antennas to cover a larger area, while still being carefully synchronised to a central unit, many unwanted effects in a MIMO system such as shadowing and spatial correlation can be mitigated. This is the idea of a distributed MIMO (D-MIMO) system. The potential of D-MIMO has been observed in other studies and testbed applications. Proving that a D-MIMO system can be reliably and accurately simulated would enable a cheap and time-effective way to further investigate the capabilities and development of new D-MIMO systems. This project investigated the similarities and differences between a real D-MIMO system and a simulated D-MIMO system by comparing simulated data to measured data obtained from a testbed. After showing that simulations can be a reliable way to study D-MIMO, further investigations were carried out. The thesis shows an overall higher capacity for a D-MIMO system compared to a regular co-located MIMO (C-MIMO) system. For an indoor office environment it is shown that similar capacity can be reached when using 12 transmitter antennas for C-MIMO as with 7 transmitter antennas for D-MIMO. Further simulations were made comparing higher mmWave signal frequencies, which still showed D-MIMO being advantageous compared to C-MIMO. However, the higher frequency simulations also showed a less stable system for both C-MIMO and D-MIMO. Lastly, outdoor simulations were made that can be recreated and confirmed with a testbed in a future project. The outdoor simulations also show a higher capacity for the D-MIMO system compared to the C-MIMO system. Overall, the results show that a D-MIMO system can be created in a simulated environment which is important for further investigating the capabilities of future DMIMO system deployments. It is also shown that the D-MIMO system is superior to an ordinary C-MIMO system with regards to received power, coverage and capacity.