Dr. Merkle, what makes the 5G network - which is not even widely available yet - already in need of improvement?

Dr. Thomas Merkle: The development of the mobile network infrastructure is a technical evolutionary process. The 5G network is not bad or inadequate, it's just that development is continuing in parallel. The big visions for the future, for example in the area of augmented reality or autonomous driving, envisage a further scaling of network capacities in the long term. This scaling can only be driven forward if higher frequency bands are also used to connect the base stations. Fiber optics are not available or usable everywhere. Broadband radio relay connections at high frequencies above 100 GHz expand the scope for network planners and allow them to control expansion in line with demand and reduce costs for subscribers. Flexible network connections in combination with autonomous and self-learning network control elements are promising key technologies here.

What goal are you therefore aiming for with the project 'Artificial
Intelligence Aided D-band Network for 5G Long Term Evolution' - Ariadne for short?

We are not trying to improve the radio network at a local level - i.e. by making selective adjustments to the network infrastructure - but are researching a longer-term solution for a reliable and energy-efficient mobile network with high transmission capacities. To this end, we also want to use frequencies in the D-band for the first time, i.e. at 130 to 174.8 GHz. These have not been used to date, but are ideal for fast data transfer. By the end of the project in 2022, we want to realize and demonstrate a stable wireless connection with extremely high data rates in the 100 Gbit/s range with almost zero latency in the D-band. The latency is mainly caused at the network control level and by the signal processing required, as electromagnetic waves themselves propagate at the speed of light - there is nothing that can be done to improve this. Forward-looking intelligent concepts can lead to a further improvement in 5G.

What are the most important tasks of the project?

The Ariadne project is based on three major complementary research areas. Firstly, the aim is to develop new radio technologies for communication in the D-band that are able to make optimum use of the frequency range. Secondly, meta-surfaces - reflective surfaces for radio waves whose reflective behavior can be electrically modified - are being researched for the high frequencies in order to overcome the lack of line-of-sight connections in urban areas. The third task area deals with network control. This is to be optimized using machine learning and artificial intelligence methods. This should enable the network control system to adapt to changes in radio conditions in the future and avoid disruptions before they even occur.

What are the project parameters?

Eleven European partners are pooling their expertise in the project. The project started at the beginning of the year and is scheduled to run for three years. It is being funded as part of the EU's Horizon 2020 research and innovation funding program with a total of around 6 million euros and is part of a portfolio of several projects that are being funded within this framework. Overall, these projects form a modular system of future technologies that are intended to support the evolutionary process of 5G.

By when do you want to have achieved which final goal?

We want to start the first real-time data transmission trials in the D-band as early as next year. The aim of the project is not just to create a transmission, but to demonstrate the reconfigurability of the connections with high spectral efficiency and low energy consumption. The development and application of reconfigurable surfaces in the D-band and intelligent network control are to be demonstrated in 2022. The project partners involved have already been able to show that the concepts are an attractive solution at low frequencies. The exciting research question now is how these concepts can be applied at frequencies above 100 GHz.