Community Research and Development Information Service - CORDIS


mmMAGIC Report Summary

Project ID: 671650
Funded under: H2020-EU.

Periodic Reporting for period 1 - mmMAGIC (Millimetre-Wave Based Mobile Radio Access Network for Fifth Generation Integrated Communications)

Reporting period: 2015-07-01 to 2016-06-30

Summary of the context and overall objectives of the project

The mmMAGIC project, co-funded by the European Commission’s 5G PPP program, is led by Samsung and brings together major infrastructure vendors (Samsung, Ericsson, Alcatel-Lucent, Huawei, Intel, Nokia), major European operators (Orange, Telefonica), leading research institutes and universities (Fraunhofer HHI, CEA-LETI, IMDEA Networks, Universities Aalto, Bristol, Chalmers and Dresden), measurement equipment vendors (Keysight Technologies, Rohde & Schwarz) and one SME (Qamcom).

The main objective of mmMAGIC is to develop concepts and key components for a new 5G mobile radio access technology which is expected to operate in a range of frequency bands between 6 and 100 GHz, namely the millimetre wave (mm-wave) frequencies. The use of such extremely high frequencies for mobile communications is challenging but necessary for supporting 5G’s extreme mobile broadband service which requires very high (up to 10 Gbps) data rates, and in some scenarios, also very low end-to-end latencies (less than 5 ms).

The mmMAGIC project develops and designs new concepts for mobile radio access technology (RAT) for deployment in the 6-100 GHz range, including novel waveform, frame structure and numerology, novel adaptive and cooperative beam-forming and tracking techniques to address the specific challenges of mm-wave mobile propagation. This new RAT is envisaged as a key component in the overall 5G multi-RAT ecosystem. Seamless and flexible integration with other 5G and LTE radio interfaces are foreseen in the design of mmMAGIC’s radio network architecture and realized through improved and entirely novel inter-networking functionalities that are being developed in the project. Self-backhauling and front hauling capabilities are studied, and is expected to create a holistic, scalable and economically viable integrated 5G solution to meet future needs of operators, enabling, for example, ultra-high definition TV and video streaming, virtual reality, immersive experience, and ultra-responsive cloud services in 5G for mobile users.

The project undertakes extensive radio channel measurements in the 6-100 GHz range at multiple locations in Europe, and is developing advanced channel models for rigorous validation and feasibility analysis of the proposed concepts and system, as well as for usage in regulatory and standards fora. The ambition of the project is to pave the way for a European head start in 5G standards and to be a focal point for European and global consensus building on the architecture, key components and spectrum for 5G systems operating above 6 GHz.

mmMAGIC is organized using a flat project structure with parallel activities taking place in different work packages. In the work packages, research is performed in relevant areas and the resulting concepts and findings will be shared and used by internal and, where applicable, external project stakeholders. The mmMAGIC work plan is structured into six work packages. The research and technology development work in the technical work packages ranges from defining user needs and implications of regulatory constraints (WP1), via channel measurements and modelling (WP2), research on system and radio interface concepts and solutions (WP3-WP5) to dissemination and visualization of results (WP6 and WP1).

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

WP1: Use case scenarios, KPIs and spectrum
The requirements and key performance indicators (KPIs) for a variety of the state-of-the-art 5G use cases, foreseen to be greatly benefited when operating in frequency bands in the range 6-100 GHz, have been defined. Three frequency blocks, including the low (6 – 30 GHz), medium (31 – 51 GHz), and high (71 – 100 GHz) ranges, have been analysed by means of four major KPIs, namely coverage, capacity, mobility, and device complexity, where the suitability of the spectrum within 6 and 100 GHz for mobile applications has been identified. A visualization tool based on the ray tracing channel demonstrating collective results on beamforming performance from the project is currently being developed.

WP2: Channel measurements and modelling
Significant efforts have been made on the framework of work carried out on channel measurements and modelling, including the preparation and validation of channel sounders, the selection of most relevant scenarios and compilation of the overall measurement plan, the preliminary evaluation of the measurement and simulation data, and the derivation of channel characteristics in different scenarios and at multiple frequencies. Based on over 30 measurements at different frequencies for six indoor/outdoor scenarios, namely the street canyon, open square, O2I, office, shopping mall and airport, an initial mmMAGIC channel model, covering frequency range from 10 to 80 GHz, has been developed and implemented as an open source software, where further extensions are currently under study based on investigations of specific aspects of mm-wave propagation such as wall reflection and diffuse scattering.

WP3: Deployment and RAN integration
The work carried out under network integration of mm-wave radio access technology (RAT) in the 5G mobile network has provided initial identification of a range of architectural aspects, including their requirements and challenges, that are crucial for such integration. The implications of network slicing to radio access network (RAN) have been discussed, and the concepts for the multi-layer protocol stack design that incorporate energy and costs metrics for optimal operation have been studied. Investigations on network integration for an edge-less mm-wave access including multi-connectivity and mobility has been carried out, where solutions for mm-wave cell clustering have been proposed to provide highly efficient mobility management and limit Core Network (CN) signalling. Methodology for obtaining backhaul requirements from use cases, deployments and predicted traffic has been studied, in addition to the key challenges for self-backhauling and resource allocation for backhaul and access, which have been analyzed in dimensions of time, frequency and space.

WP4: Design of mm-wave mobile radio interface
Extensive and in-depth research has been carried out in the air interface topics at mm-wave frequencies. In particular, state-of-the-art waveforms have been analyzed and simulated using different KPIs and a common simulator among all partners contributing to this work. Three families of channel codes, namely the Turbo code, low density parity check (LDPC) codes, and Polar codes, have been studied for a number of performance measures, and it has been concluded that, although the use of single code family is preferable, a channel code solution for mm-wave RAT can consist a combination of more than one code family in order to fulfil diverse use case requirements. Novel reliable and efficient HARQ methods have been proposed, where their necessity for above 6 GHz communication is envisaged from the study. Frame structure and numerology design has been looked into considering KPIs such as throughput, robustness, latency and mobility support, as well as design principles targeting at different use cases and requirements. Four subframe structures, with the possibility of free configuration to enable flexible time division multiplexing, have been proposed, in addition to one specific frame structure to enable efficient backhauling operation, by allowing a specific resource for backhaul uplink and downlink, and access downlink.

Requirements and general design principles under mm-wave propagation for air interface on multiple access and duplexing, and initial access schemes have been studied, where the use of massive antenna arrays, and the support of different antenna array configuration and hardware architecture have been taken into account. Such studies have led to novel algorithms and mechanisms for efficient spectrum sharing and beam scheduling, and novel initial access schemes under the framework of broadcast signaling, sweeping subframes, beam codebook design and Quality of Service (QoS)-centric resource allocation for multi-UE access, which have been proposed, modelled, and evaluated. Hardware-in-the-loop (HIL) experiments that cover the performance tests of signal waveforms and beam steering algorithms in the presence of hardware impairments and/or influences of real world environment are currently being developed.

WP5: Multi-node and multi-antenna transceiver architectures and schemes
Hardware impairments such as the phase noise, power amplifier non-linearity, In-phase and Quadrature phase (IQ) imbalance, and antenna imperfections under mm-wave have been developed. The developed phase noise model has been made to public as an open source code, where the behaviour of wideband antenna arrays (up to 64 elements) at 28 GHz was parameterized and combined with the channel model developed in the project, enabling the generation of directional channel responses at the given frequency band. Initial multi-antenna analogue, digital, and hybrid beamforming solutions have been studied, where wideband hybrid beamforming schemes using a subset of sub-carriers for beamsteering have been developed. Coordinated mm-wave and LTE Macro cell backhaul solutions have been proposed for 5G moving hotspots, and effective relay node selection schemes have been studied for mm-wave deflection routing in dense deployments. A number of work has also been carried out on advanced multi-node cooperation and coordination, including the initial assessments on the number of line of sight inks available to a typical outdoor mm-wave user, as well as the possibility of combining mm-wave and free space optic links to provide diversity gains. Finally, mm-wave coverage maps based on real map data and ray tracing have been developed to analyze the impact of the number of mm-wave access points on the overall outage probability.

The project produced the following results in the first year:
• Two mmMAGIC workshops were held with the Advisory Board to share project result with stakeholders with regulatory, standards, and wider user groups. The captured feedback was fed into the project’s technical work, including a mode detailed action plan, based on the latest regulatory and standards development.
• Coordinated dissemination of results to 3GPP via industrial partners, resulted in thirteen contributions on channel modelling above 6 GHz that have highlighted or referred mmMAGIC, and one contribution related to hardware impairments modelling.
• Coordinate dissemination of results to ITU-R and CEPT via industrial partners.
• Organized dissemination of results to the spectrum, pre-standards, and architecture Working Groups of 5G Infrastructure Association and also contributed to the cross-project white papers along other 5G PPP projects involved in these Working Groups.
• Coordinated dissemination of mmMAGIC results to a wider scientific community via all partners. So far mmMAGIC has delivered a portfolio of work to the scientific community which is briefly listed below:
• 43 scientific conferences and workshops have been submitted, accepted or published along key IEEE conferences and industry events.
• 10 journals submitted or accepted in reputable transactions, IEEE magazines and letters.
• 17 invited talks, panels and poster presentations along key workshops and conferences and EU telecom industry events.
• 10 organized / co-organized (along other 5GPPP/ EU projects) workshops, special sessions and panels along key conferences.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

From the technical perspective, the study carried out on identifying the suitability of the spectrum band in the 6-100 GHz range for 5G systems, in particular, by considering predominate key 5G requirements including capacity, coverage, mobility and device complexity, has characterized representative candidate bands, and provided a concrete and comprehensive measure of the wide mm-wave spectrum bands that are available for a variety of mobile applications supported by 5G. Such results are of importance to maintain discussions with other parties such as regulators and 5G PPP Spectrum Working Group (WG).

In addition, the channel measurement campaigns in multiple 5G deployment scenarios over 6GHz has contributed to close the knowledge gaps for channel modelling at mm-wave band. Multi-frequency measurements with omnidirectional and directional antennas for different environments, and the dedicated measurements on human blockage, wall diffuse scattering, traffic impact and channel sounder validation, have contributed valuable state-of-the-art (SoTA) in the channel measurements and modelling for 5G mm-wave bands. The mmMAGIC initial channel model is made available to the public as an open source simulator, which is expected to be used in a wide research and technology community for evaluation and design of 5G mm-wave systems.

At mmMAGIC we have identified a range of architectural aspects that are crucial for the integration of mm-wave RAT in 5G mobile network. Several multi-connectivity solutions in the RAN architecture level have been proposed which will affect the ongoing and future Control- and User-Plane design. The project innovates on solutions for mobility management using cell-clustering, multi-RAT management and energy efficiency. These solutions are timely relevant for their potential to be implemented as an extension of LTE Rel-12 Dual Connectivity.

A big step beyond the SoTA from the air interface perspective is the validation and assessment of air interfaces considering mm-wave specific challenges such as strong hardware impairments and low power efficiency of amplifiers, as well as taking into account the possible high Doppler spread, strong phase noise, and the large amount of available bandwidth. The proposal of multiple access and duplexing schemes considering the extensive use of beamforming and spatial multiplexing opportunities has contributed novel access schemes considering the unique directional transmission feature of mm-wave communication. New multi-access mechanism has been introduced and shown to have solved the main problems of the current state-of-the-art protocols.

The innovative phase noise model has contributed to public as open source codes, and the power amplifier (PA) non-linearity has been modelled using a novel behavioural and statistical approach. The antenna imperfection modelling and parameterisation at 24.25 – 27.5GHz have enabled directional channel models to be created. These outputs greatly facilitates the evaluation of the performance and novel designs of 5G mm-wave communications, especially in the presence of component imperfections and directional antennas at mm-wave band, and have supported to push the knowledge beyond SoTA in these areas.

Finally, in-depth research is being carried out on multi-antenna and multi-node coordination, where insightful initial results taking into account of effects of, for example, the highly dense mm-wave network and possibly behaviour change of the interference due to beamforming, have been made available in the first year, and has contributed to bring the current theoretical findings and research in multi-node coordination a step further.

With the technical results achieved in the first year, mmMAGIC has showed great impact in the development of new mm-wave networks on globally accepted standards, such as 3GPP, by having 14 contributions to 3GPP RAN 1 that has been acknowledged or cited the mmMAGIC work. mmMAGIC is also actively engaged to the assessment and harmonization of suitable spectrum for 5G above 6 GHz, via contributions to Working Party 5D, CPG and 5GPPP Spectrum WG where several industry members of the consortium (Samsung, Ericsson, Alcatel-Lucent, Intel, Huawei, Nokia) are active contributors, as well as via actively engaging European regulators (Germany, France, UK, Sweden and Finland) in the advisory board.

The knowledge gained in the collaborative research carried out in mmMAGIC greatly boost the technology readiness of the industry and SME partners for 5G as well as significantly improving their capacity for innovation and competiveness towards new products and services. The European headquartered communication networks vendors in mmMAGIC (Ericsson, Alcatel-Lucent, Nokia) have together a very large global market share, and the innovation achieved within mmMAGIC is expected to give them a head-start in the race to 5G systems operating in high frequency bands, thereby helping them to improve their future share of 5G infrastructure. Results from mmMAGIC on, e.g., system integration, network architecture, and cost and energy efficient hardware, will help operators to define an evolution strategy towards 5G and to avoid short term decisions that may preclude or impair the adoption of mm-wave based network solutions. From mmMAGIC, the operators obtain realistic and quantified assessments of the maturity of the different technological components that will be required in order to have the whole system available, facilitating also the identification of cooperation opportunities with vendors to help speed up their development. In the longer term, our system will enable many of the foreseen 5G services and vertical applications which require large bandwidth, provide superior quality of experience for the end user (higher data rates, ubiquitous availability, very low delays), hence creating significantly new revenues streams for operators and service providers.

mmMAGIC project is one the RTD projects in the phase 1 of the 5G PPP Initiative. The project is interacting with other concurrently running 5G PPP projects. The Coordinator and the Technical Manager of mmMAGIC, who represent the project in the 5G PPP Initiative’s Steering Board and Technology Board, respectively, provides means for closed collaboration between different 5G PPP projects. This form of close collaboration has ensured that technology concepts and solutions in mmMAGIC are being developed as an integral component of the holistic 5G PPP approach, and consequently impact the direction and outcomes of other related projects, e.g., projects related to 5G system and network architectures, other 5G RATs, and hardware components. The outcomes of the mmMAGIC project regarding, e.g. the feasibility of the developed systems, has been fed into this process, through project contributions to 5G Infrastructure Associations Working Group on 5G Vision, thereby helping to shape and impact strategic research priorities of the subsequent phases of 5G PPP, including mm-wave communications.

Last but not least, mmMAGIC disseminated over 50 scientific publications, hosted a number of panels, and organised workshops at high standards conferences such as VTC, GLOBECOM, EuCAP, PIMRC, that has impacted a wider research community and academy. The project provides a link between academy and industry from theory to practice. The results from the project have also been shaping new academy courses such as: course on future wireless communications at Aalto University (Finland), course on 5G Communications and Technologies at University of Surrey (UK), and another course on “Challenges and Opportunities with mm-wave Communications in 5G” in the Master Program on Communication Engineering (MPCOM) at University of Chalmers (Sweden).

Related information

Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top