5G is superior to its predecessors, but in urban areas where direct visual contact between the transmitter and receiver is difficult, the radio connection often does not work reliably. A case for the ARIADNE research project, in which Fraunhofer IAF is also involved.
Currently building the fifth generation of mobile communications, research is already working on optimizing it. The great advantages of 5G are the high frequencies and the associated high transmission rates, which ensure an almost latency-free connection and fast data exchange. However, the high frequencies are accompanied by a directional system that usually requires a line-of-sight, or LOS for short: transmitters and receivers must see each other. But the principle of the LOS connection could lead to connection problems, especially in urban and very densely built-up areas.
One of the problems that lead to these connection problems in local 5G networks is the wiping effect. This effect occurs when a signal is transmitted via a line of sight and at the same time copied via reflections. The copy overlay the signal from the line of sight and delete it. The result: the signal does not arrive at the receiver. This multi path propagation via a non-line connection (non-line-of-sight, NLOS for short) remains a problem with 5G, just like with its predecessor 4G. For this reason, one of the main goals of ARIADNE is the development of new concepts that better master these LOS and NLOS scenarios and massively increase the reliability of mobile phone connections.
In the recently launched EU project with the full title “Artificial Intelligence Aided D-band Network for 5G Long Term Evolution”, eleven partners from five countries are researching how the use of high-frequency bands and artificial intelligence can lead to an advanced system architecture for Beyond 5G Let develop.
Greater Efficiency And Reliability of 5G
The aim is to develop energy-efficient and reliable mobile communications based on frequencies in the D band (130-174.8 GHz). Since the D-Band has an aggregate bandwidth of more than 30 GHz, it is ideal for fast data traffic. However, this newly used band is divided into several sub-bands and is therefore not contiguous, which necessitates an adaptation of the radio system architecture used previously and the corresponding network control.
In ARIADNE, the combination of a new high-frequency radio architecture and a new network processing concept based on artificial intelligence is intended to create an intelligent communication system “Beyond 5G”, that is, as a further development of 5G. By 2022, the project consortium wants to implement and demonstrate a radio link with extremely high data rates in the 100 GBit / s range with almost zero latency.
ARIADNE is based on three major research areas: the development of hardware components, the research of meta-surfaces, and the adaptation of network control based on artificial intelligence or machine learning.
Components For A Reliable D-Band Connection
Fraunhofer IAF brings its expertise from the field of high-frequency electronics to the development of hardware components: together with its partners, the Freiburg researchers are developing new radio technologies for communication in the D band (130-174.8 GHz). “Our focus is on the development of new radio modules with the highest spectral efficiency that takes advantage of frequency diversity and offers a control interface for optimization in the network.
To avoid network disturbances in NLOS connections, meta surfaces (meta surfaces) and their contribution to an optimization of the radio connection are researched in ARIADNE. Meta surfaces are adjustable reflectors for radio waves and are intended to counteract network control problems in urban areas. If there is no line of sight between base stations on the roofs of the house and the users in the house canyons, meta-surfaces should reflect radio waves and thus ensure the spread outside the line of sight. The control of the meta surfaces should take place via a central network controller.
The concept of metasurfaces is already partially implemented for 5G, but so far only for low frequencies. The higher the frequency of the radio connection, the finer the microstructures on the surface must be, and for frequencies in the D-band, the structures are very complex to manufacture, explains Thomas Merkle. For this reason, the project team is researching the development of metasurfaces that are suitable for both high frequencies and industrial production. At Fraunhofer IAF, researchers are working on the so-called reflectarrays. These are small meta-surfaces on antennas that are used for beam swiveling and focusing.
AI-Based Network Control
To provide a constant and reliable radio connection in all weather conditions, methods of machine learning and artificial intelligence (AI) should be used for network management. So far, mostly classic mathematical methods have been used for mobile radio control. ARIADNE will now use AI-based algorithms to solve radio communication problems. While thorough data analysis is the goal in machine learning, AI is to be used to develop a system for network control that not only recognizes and reacts to problems but can even predict and prevent them.
The final goal of the project partners is to bring the individual project modules together in a test system and to demonstrate the functionality. At the end of the project, they want to present two demonstrators as a result of their research work: The first demonstrator should achieve a connection over 100 meters that is reliable in all weather conditions with a data rate of 100 GBit / s. As a proof-of-concept in the laboratory, the second demonstrator is intended to show how a meta-surface can improve the propagation conditions for radio transmission in the environment. This is how the functionality of meta surfaces at high frequencies is to be proven in the laboratory.