Multi-antenna configuration modeling for massive MIMO V2I

Vehicle communications are now emerging towards the 5th Generation Wireless Systems (5G). Shortly, an explosive increase in the number of connected vehicles is expected. This communication is not only for car to car safety reasons, but also for communicating with the environment concerning entertainment and internet. Larger data rates as provided by current Long Term Evolution (LTE) systems are required. Energy efficiency is becoming more and more important as part of the green communication for the 5G network. Consequently, there is a growing research interest in massive Multiple Input Multiple Output (MIMO) antenna systems providing higher channel capacity, spectral and energy efficiency compared to conventional MIMO without beamforming. Also conventional MIMO needs to be investigated for Vehicle to Infrastructure (V2I) communication, since the evolution towards the 5G network is slow hence interim solutions are essential. Furthermore, the impact on the channel capacity for different inter-element spacing for antennas on top of the car is examined. In order to study the impact of different multi-antenna geometries and MIMO modalities in both vehicle and fixed Base Station (BS), a quality modeling of a realistic scenario is required. The analysis is based on a numerical simulated model of an urban scenario, emulating as a case of study a portion of the city of Barcelona in a realistic environment, comprising of a massive MIMO base station and a car at various positions on a trajectory. Several systems and configurations are compared by means of the channel eigenvalues and capacity. On top of the vehicle one, two or four monopole antennas are used for the investigation. Corresponding at the BS one, two or four patch antennas are used for conventional MIMO and 64 patch antennas grouped as one, two or four beams for massive MIMO. The following key findings are found for V2I communication in the resented work. For all massive MIMO configurations the interference for other users is highly reduced. Furthermore, since there is a huge increase of received power for the vehicle without losing the heterogeneity of the eigenvalues, this leads to high channel capacity with less transmitted power. For a Signal to Noise Ratio (SNR) of 12 dB with conventional MIMO 4x4 almost 170% more average channel capacity can be gained with respect to a Single Input Single Output (SISO) in Line of Sight (LOS). For Non Line of Sight (NLOS) it is even 190%, which is very close to the ideal limit. For different inter-element spacing, the channel capacity also depends on the amount of elements in the car. The higher the number of elements, the higher is the achievable channel capacity, but larger inter-element distance is necessary to obtain the optimal performance.