maximumMIMO - Maximum spectral Efficiency by parallel Multiple-Input-Multiple-Output Transfer using high density 3D-Antenna-topology

Proposer

Prof. Dr.-Ing. Gerhard P. Fettweis

Technical University Dresden

Institut für Nachrichtentechnik

Vodafone Chair Mobile Communications Systems

Prof. Dr.-Ing. Eckhard Graß

Humboldt-Universität zu Berlin

Institut für Informatik

Prof. Dr.-Ing. Berthold Lankl

Universität der Bundeswehr München

Fakultät für Elektrotechnik und Informationstechnik

Institut für Informationstechnik (EIT 3)

Abstract

In recent years, data rates of wireless communications systems have increased by a factor of ten every five years. Applications like wireless backhaul in cellular networks will soon require data rates of 100 Gb/s, which is at least one order of magnitude away from what is feasible today. Therefore, it is high time to start research activities paving the way towards wireless data rates of 100 Gb/s and beyond. There are several scientific challenges involved in achieving 100 Gb/s. Two of the main challenges arise from limitations of the hardware that is available to build practical wireless systems. That is, these limitations confine the frequency bandwidth that can be used, and they confine the speed of the serial signal processing that is essential for wireless data transmission. One approach to overcome these issues is to design wireless systems with ultra-high bandwidth efficiency and a large degree of inherent parallelization for the signal processing. This approach is considered in the project “maximumMIMO”, which jointly addresses two of the research areas of interest of the DFG priority program 1655, namely “baseband design for extremely high spectral efficiency” and parallelized “system architectures for extremely high data throughput”.

Initial link budget calculations have shown that it could be reasonable to use a very high number of transmit and receive antenna-elements rather than a high order modulation scheme to maximize the spectral efficiency. Thus, multiple-input multiple-output (MIMO) transmission with many-element antennas appears to be very promising for wireless data rates of 100 Gb/s, which is the research focus of “maximumMIMO”. So far, there has been the general notion that wireless MIMO systems can achieve spatial multiplexing gains only in non-line-of-sight environments. Recent research has proven, however, that spatial multiplexing is even feasible under line-of sight (LOS) conditions. Furthermore, it has been shown that 3D antenna topologies may be better suited for this purpose than conventional planar antenna arrays.

Therefore, “maximumMIMO” focuses on a theoretical characterization and the design of LOS MIMO systems with optimal 3D many-element antennas. The research is of general nature and applicable to any frequency band. For practical experiments and hardware demonstrations, 60 GHz technology is used, which is suited, e.g., for wireless backhaul links in cellular networks.

The research of “maximumMIMO” covers 3D LOS MIMO channel measurements and modeling, information- theoretic analyses of optimal antenna topologies under hardware und robustness constraints, as well as design and implementation concepts for highly-parallelized MIMO signal processing, synchronization, medium access control and network interface functionality. After a general proof of concept in the first project phase, an extension to more general 3D MIMO channels and the design of a full 100 Gb/s demonstrator are planned (second phase).