BI-SDMoF: Bit-Interleaved Sigma-Delta Modulation over Fiber
Description of the PI
Piet Demeester is full professor in the Department of Information Technology (INTEC) of the Faculty of Engineering and Architecture at Ghent University (UGent). He is leading the activities of the IDLab research group (www.idlab.ugent.be) (idlab.technology) that is also a department in the strategic research centre imec. IDLab performs fundamental and applied research on internet technology and data science.
Major research areas are machine learning and data mining; semantic intelligence; distributed intelligence for IoT; cloud and big data infrastructures; multimedia processing; wireless and fixed networking; electromagnetics, RF and high-speed circuits and systems. IDLab has a unique research infrastructure used in numerous national and international collaborations.
Piet Demeester is chair of the ERC consolidator panel “PE7–Systems and Communication Engineering” (2013-2015-2017). He is associate editor of the IEEE Journal of Lightwave Technologies and of the IEEE/OSA Journal of Optical Communications and Networking. Piet Demeester is Fellow of the IEEE and holds an ERC advanced grant (2017-2021, www.atto.ugent.be).
Description of the project
Network operators are struggling on how to release broadband mobile services in highly dense and hot-spot scenarios. The 5th generation of mobile networks is targeting per user downlink and uplink rates of 300 Mb/s and 50 Mb/s respectively. In ultra-dense environments such as stadiums, airports, shopping malls and tourist hot-spots, the aggregated bit rate becomes enormous. To make it more concrete, Belgium’s national football stadium (King Baudouin) is taken as example. The stadium can accommodate 50.000 spectators. Even if an average bit-rate of only 50 Mb/s needs to be provided, an aggregated bit-rate of 2.5 Tb/s is required. Given the small area of a stadium (18.000 m2 seating area), this results in an astonishing capacity per area of 140 Tb/s/km2 or 140 Mb/s/m2. This is a 10-fold increase compared to area traffic capacity targeted in 5G and a 100-fold increase compared to the current 4G technologies.
To go beyond the capacity per area envisioned in 5G and support very high bit rates in ultra-dense environments, MaMIMO technology should be exploited at mm-wave frequencies (typ. 28 and 60 GHz range). However, this comes with an additional difficulty: MaMIMO demands a very rich scattering environment to support a very high number of orthogonal streams. However, the absorption at mm-wave frequencies limits scattering. To overcome this, a distributed antenna system (DAS) is proposed. Instead of relying on passive scattering, different antennas can be distributed to mimic reflections actively. As such, the rich scattering environment at sub-6GHz frequency can be emulated at mm-wave frequencies. However, very tight synchronization of the transmitted signals is required, which requires full centralized processing, resulting again in the bandwidth explosion of CPRI. Alternatively, analog radio signals can be transported over fiber. Notwithstanding, this is much more bandwidth efficient, it is also expensive: very linear electro-optical modulators are required. Furthermore, with analog RoF (ARoF), signals are transported as IF signals multiplexed over different frequencies. To enable transmission of the radio signals, additional filtering and the generation of different local oscillators is required.In BI-SDMoF, we propose the concept of a bit-interleaved sigma-delta modulation over fiber to distribute the radio signals in a more efficient way.
