15 Methodology for Modeling the Energy …
235
resulting in an efficient spatial filtering. The receiver benefits from the signal amplification effect and improved interference suppression. In this way, beamforming
further assists spectral efficiency of Massive MIMO and reduces also transmission
power consumption.
But Massive MIMO and beamforming have also some potentially negative effects
in respect to absolute power consumption and material use. Beamforming is a signal
processing procedure creating individual amplitude or phase variations at the transmitter. The mechanism used for beamforming (analog, hybrid, or fully digital) has
an impact on energy consumption (Rohde and Schwartz 2016). Depending on the
system design and scale of the antenna array, system energy consumption varies (Ali
et al. 2017; Roth and Pirzadeh 2017). The Massive MIMO and beamforming concept
means that each antenna or every second antenna of the array has a dedicated power
amplifier and radio signal processing (RF-chain). In comparison, in current LTE
antenna modules one RF-chain is allocated to 8 physical antennas. The much larger
number of RF-chains in Massive MIMO will influence not only the total system
energy consumption, but the material use as well. Conventional silicon CMOS transistors have limits with regards to high frequency (mmWave) signal processing and
amplification. Future high speed and low power devices require higher bandgap III–
V semiconductors such as gallium nitride (GaN), gallium arsenide (GaAs, silicon
germanium (SiGe), or indium phosphide (InP). The group III elements gallium,
germanium, and indium are labeled by the European Commission as critical raw
materials (CRM List of European Commission 2017). The group V elements have
some toxic potential.
The research question in this context is; what is the best beamforming solution
for cmWave and mmWave Massive MIMO antenna arrays?
15.5.3 Small Cells and Network Densification
The last technology concept for improving data traffic capacity is to increase the
number of bases stations or antenna sites. This densification of the radio access
network is absolutely necessary when utilizing new mmWave spectrum. In Europe
the pioneer spectrum and most likely frequency band is 26 GHz. But even in the
already auctioned 3.6 GHz spectrum a trend to smaller cell sizes is likely to be seen
in order to provide higher data rates to a single user.
The analysis and assessment of the environmental impact of this network densification is complex. There are several factors that increase and decrease the resulting
energy and material consumption. In current networks, the main driver for energy
usage is the transmission of signals over longer distances from antennas to users.
With 5G’s smaller cells, less energy is needed for transmission, but computation
power increases due to more active antennas and complex signal processing. While
it is feasible to assume that for a single antenna system the energy and resource
consumption will decrease, the considerably larger number of active antennas as
well as antenna sites per area, the absolute energy and material consumption could
235
resulting in an efficient spatial filtering. The receiver benefits from the signal amplification effect and improved interference suppression. In this way, beamforming
further assists spectral efficiency of Massive MIMO and reduces also transmission
power consumption.
But Massive MIMO and beamforming have also some potentially negative effects
in respect to absolute power consumption and material use. Beamforming is a signal
processing procedure creating individual amplitude or phase variations at the transmitter. The mechanism used for beamforming (analog, hybrid, or fully digital) has
an impact on energy consumption (Rohde and Schwartz 2016). Depending on the
system design and scale of the antenna array, system energy consumption varies (Ali
et al. 2017; Roth and Pirzadeh 2017). The Massive MIMO and beamforming concept
means that each antenna or every second antenna of the array has a dedicated power
amplifier and radio signal processing (RF-chain). In comparison, in current LTE
antenna modules one RF-chain is allocated to 8 physical antennas. The much larger
number of RF-chains in Massive MIMO will influence not only the total system
energy consumption, but the material use as well. Conventional silicon CMOS transistors have limits with regards to high frequency (mmWave) signal processing and
amplification. Future high speed and low power devices require higher bandgap III–
V semiconductors such as gallium nitride (GaN), gallium arsenide (GaAs, silicon
germanium (SiGe), or indium phosphide (InP). The group III elements gallium,
germanium, and indium are labeled by the European Commission as critical raw
materials (CRM List of European Commission 2017). The group V elements have
some toxic potential.
The research question in this context is; what is the best beamforming solution
for cmWave and mmWave Massive MIMO antenna arrays?
15.5.3 Small Cells and Network Densification
The last technology concept for improving data traffic capacity is to increase the
number of bases stations or antenna sites. This densification of the radio access
network is absolutely necessary when utilizing new mmWave spectrum. In Europe
the pioneer spectrum and most likely frequency band is 26 GHz. But even in the
already auctioned 3.6 GHz spectrum a trend to smaller cell sizes is likely to be seen
in order to provide higher data rates to a single user.
The analysis and assessment of the environmental impact of this network densification is complex. There are several factors that increase and decrease the resulting
energy and material consumption. In current networks, the main driver for energy
usage is the transmission of signals over longer distances from antennas to users.
With 5G’s smaller cells, less energy is needed for transmission, but computation
power increases due to more active antennas and complex signal processing. While
it is feasible to assume that for a single antenna system the energy and resource
consumption will decrease, the considerably larger number of active antennas as
well as antenna sites per area, the absolute energy and material consumption could