RAPID

Radio technologies for 5G using Advanced Photonic Infrastructure for Dense user environments

In RAPID, we propose to use a centralized radio access network (C-RAN) architecture to support high-capacity heterogeneous (3G, 4G and 60 GHz) radio access technologies through low-cost, but ultra-high-bandwidth photonic techniques for the fibre distribution. To meet the requirement of transporting the high-frequency (and wide-bandwidth) wireless signals at a low-cost, a novel coherent radio-over-fiber (CRoF) scheme is proposed in RAPID. Furthermore, for the mobile receiver, novel extremely low-cost (<10€) integrated transceivers will be developed based upon SiGe technology.

The radio resource management of the heterogeneous network using the developed mm- wave and photonic technologies together with legacy 3G/4G wireless will be demonstrated in RAPID and the developed hardware will be tested in life-cycle assessments within different real life networks scenarios including a public stadium, an operator’s fiber-optic and wireless network, as well as in train, airport scenarios

List of participants

Participant's role Organisation name Country
EC-Coordinator Universität Duisburg-Essen Germany
  University of Kent United Kingdom
  Corning Corporation Germany
  Siklu Inc. Israel
  Exatel Poland
JP-Coordinator Osaka University Japan
  Doshisha University Japan
  Electronic Navigation Research Institute Japan
  Hitachi Ltd. Japan
  Central Research Institute of Electric Power Industry Japan
  Koden Techno Info Japan

Key objectives

Future radio/wireless access in dense user environments, with the high bit-rates envisioned per user for 5G communications and the increased numbers of users expected to request such high bit-rate services, requires shortening the wireless propagation distances and, hence, the use of a fibre optic infrastructure to the wireless access points. The use of new spectrum will be important in providing the bandwidth needed, but spectrum efficiency and control of the spectrum usage according to application will still be important. Millimetre-wave frequencies, such as those around 60 GHz, offer high potential for reuse, and can thus promote spectral efficiency. The ease of beam- forming, and its significance for reasonable transmission distances (with low emitted power levels) is also important for millimetre-wave communications. For the fibre distribution to the access points, several possibilities exist. Baseband (e.g. multi-Gb/s Ethernet transport) can be used between fully functional access points, such as those being developed for millimetre-wave WLAN standards (802.11ad/aj). However, there is a general lack of coordination possible between such access points, and further spectral efficiency and throughput improvements, for example through the use of joint processing, interference cancellation or virtual MIMO techniques are not possible. Access points which are instead simple remote antenna units (RAUs) can be more flexibly coordinated by centralised baseband processing units where joint processing of signals can also be carried out. Such architectures (generally, without joint processing) have been used in indoor distributed antenna systems at lower, RF/microwave frequencies, and are being proposed for cloud- or centralised-Radio Access Networks (C-RANs).

In RAPID, we propose to use a C-RAN architecture to support high-capacity heterogeneous millimetre-wave and 3G/4G RF/microwave radio access technologies through low-cost, but ultra- high-bandwidth photonic techniques for the fibre distribution.

The lower frequency 3G/4G radio access technology can provide more general coverage but with lower bit-rates. The mm-wave radio access technology provides more localised coverage but with ultra-high-bit-rates. Coverage can be enhanced by placing more millimetre-wave enabled RAUs, or by using steered directive beams from the RAUs which enable the communication with particular users. Additional flexibility may be added by allowing devices to communicate directly with each other. The important aspect, however, is that both the millimetre-wave communications (and its spectrum use) and the legacy 3G/4G system are controlled centrally by the network infrastructure. This allows for the most efficient use of spectrum and for energy consumption optimisation within the constraints of providing the required Quality of Service/Quality of Experience (QoS/QoE) to users.
Enabling millimetre-wave cellular systems requires properly dealing with the channel impairments and propagation characteristics of the high frequency bands (larger path loss and greater noise power due to the use of wider bandwidths). With the small wavelength, arrays of practical dimensions can house orders of magnitude more elements than microwave arrays and can provide enough array gain to overcome path loss and ensure high SNR at the receiver.
Dynamic resource allocation (spectrum, power, bit rate) within 60 GHz systems is challenging because of the much larger bandwidth. Depending on the beam-forming architecture, beam-forming weights required to form the directive beam could be applied in the digital or analogue domain. Digital beam-forming is done in the form of digital pre-coding that multiplies a particular coefficient to the modulated baseband signal per RF chain. For analogue beam-forming, on the other hand, complex coefficients are applied to manipulate the RF signals by means of controlling phase shifters and/or variable gain amplifiers. When combined with an OFDM system, digital beam-forming is carried out on a subcarrier basis. Opportunistic beam-forming is a crucial element in efficient wireless spectrum resource utilisation and can be carried out by considering the channel quality of individual subcarriers in the OFDM wireless system.

In RAPID, we will compare digital and analogue beam-forming techniques to optimize link performance and reduce overall energy consumption.

In addition to investigating dynamic resource allocation within the millimetre-wave system, optimizing resource allocation between the millimetre-wave and RF/microwave systems is also important. One approach is a kind of virtual networking. That is to use the RF/microwave radio access technology (RAT) for control signalling in macro-cells, while the millimetre-wave RAT is used for data transport in small cells. This can not only increase capacity/throughput but can also reduce the energy consumption of the heterogeneous networks. This may be useful for enabling the control of high-bit-rate mm-wave downlink transmission, for example for HD/SHD video streaming, through the control of the legacy 3G/4G network.

The radio resource management of the heterogeneous network using the developed mm- wave and photonic technologies together with legacy 3G/4G wireless will be demonstrated in RAPID.

For transporting the radio signals between RAUs via optical fibre, digitized radio-over-fiber (DRoF) techniques are currently preferred in mobile networks, whereas both digitized and analogue radio signal transport have been used in commercial indoor distributed antenna systems. With the move to wider bandwidths in 4G and 5G systems, digitised radio places huge demands on bit-rates – multi-Gb/s for each c. 100 MHz-wide channel, correspondingly more for GHz-wide channels at mm-wave frequencies.

To meet the requirement to transport high-frequency (and wide-bandwidth) signals at low- cost, a novel coherent radio-over-fiber (CRoF) scheme employing photonic up-conversion techniques is proposed in RAPID. For the mobile receiver, novel extremely low-cost integrated transceivers, based on SiGe chip technology are proposed.

Furthermore, as mm-wave coverage is more localized, leading to pico- or femto-cells, a high number of cells needs to be connected. By seamless integration into standard optical WDM technology, low-cost can be ensured and also multi-band and multi-service operation is possible. The centralised control will furthermore enable capacity reallocation or service switching features.

Consequently, in RAPID we propose to connect the pico-/femto-cells seamlessly to a WDM optical access infrastructure with a centralised coordination.

The low-cost coherent photonic technologies proposed may also be of benefit to digitized radio transport over the fibre (although the bandwidth requirements will always be higher than for the analogue transport). However, the resilience to impairments of the digitized transport is attractive, and studies will be made of the application of the photonic techniques to digitized radio signal transmission and reception.

The technologies developed in RAPID will be evaluated in a number of pre-defined scenarios with densely located users including train, aircraft, stadium, school, conference, etc. expected in the scenarios of the call. Life cycle test will be carried out at sites of the RAPID beneficiaries but the RAPID consortium is also committed and has pre-arranged to conduct further life-cycle tests at additional sites in close collaboration with external project partners.

Thus, life-cycle assessment of the developed hardware in different real life networks will be a key objective in RAPID.

Further information and contact: Andreas Stöhr

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 643297.