Hardware Efficient Multi-antenna Transceivers

• Applicants: Professor Dr.-Ing. Georg Fischer; Professor Dr.-Ing. Ralf Reiner Müller
• Subject Area: Electronic Semiconductors, Components and Circuits, Integrated Systems, Sensor Technology, Theoretical Electrical Engineering
• Term: from 2018 to 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 409561515

This project aims to increase the energy efficiency and to reduce the radio-frequency (RF)-cost of multi-antenna transceivers for the fifth generation of wireless networks (5G). Novel energy- and cost-efficient schemes for multi-antenna transceivers will be studied by means of analytical tools from information theory, RF circuit theory, statistical physics, and also with the help of RF software tools. This project encompasses hardware architectures with high power efficiency and low RF-cost, as well as signaling methods allowing for hardware efficient multi-antenna transceivers. Various schemes of Hybrid Analog-Digital (HAD) transceivers for millimeter wave (mmWave) and microwave frequency bands found in literature will be compared and their fundamental limits will be investigated. In addition, novel HAD architectures will be investigated with particular emphasis on reflect arrays. HAD architectures will be analyzed by appropriate analytical and RF simulation tools. Both massive multiple-input multiple-output (MIMO) systems below 6 GHz and mmWave communication will be considered utilizing appropriate channel models. The project results will be some new hardware architectures for multi-antenna transceivers for both base station and user terminal applications with improved energy efficiency and reduced RF-cost compared to the state of the art. Furthermore, analytical and simulation results of such architectures will be provided.Along with hardware architectures, precoding and signaling methods allowing for hardware efficient multi-antenna transceivers including constant envelope and peak power-limited pre-coding will be analytically investigated by invoking the replica method from statistical physics. Furthermore, this method of analysis will be used for precoding with antenna selection in multi-antenna base stations to predict the performance of optimal antenna selection. Finally, novel low-complexity algorithms for these precoding schemes will be proposed for implemention in 5G.

• DFG Programme: Research Grants