The need for an efficient integration at multiple levels (technology, types of cells, and access-backhaul), requires the
use of several techniques that encompass the PHY layer up to the network. This triple layer integration provides
the unicity of this MASSIVE5G proposal. While the various techniques are currently being researched, the typical
approach is to consider the different aspects in isolation. Typically the access is studied independently of the backhaul
or using very crude models. Interference management (based on coordination and cooperation), cognitive radio,
mmW communications and massive MIMO are currently in most cases studied in isolation. While the partitioning
of the problems ensures a very deep analysis, when it comes to the integration it is verified that unforeseen
but very significant issues, make the system performance fall well below the promises if not compromised at all.
MASSIVE5G will consider the separation of the control and data planes, where the control plane uses the sub-6 GHz
band and the data plane both the mmW and sub-6 GHz bands. This allows to provide a highly reliable feedback
that will enable an accurate precoding design ensuring high throughput intended for the data transmission. In the
collaborative MASSIVE5G project we will go beyond the theoretical studies by implementing/evaluating the most
promising developed algorithms in a FPGA and integrating them in a radio/optical infrastructure ORCIP (Optical Radio
Convergence Infrastructure for Communications and Power Delivering) implemented in IT.
The main technical objectives are to develop a high-performance heterogeneous access network by considering
a flexible C-RAN based architecture, and integrating mmW communications with conventional radio-wave
communication systems, joint cognition and cooperative schemes, efficient and joint cross-layer approaches in the
management of radio and FH/BH resources. To reach the above main goals, specific research objectives are:
-Design and evaluate new interference management (based on pre-coding for the downlink and cooperative
equalization for the uplink) and joint scheduling mechanisms to mitigate interference through cooperation. These
will be enhanced with the knowledge of the interference obtained through spectrum sensing and cognitive radio
-Development of new methods for reducing the load in the BH/FH. This includes new methods for the acquisition/
dissemination of the information required for coordination / cooperation as well as compression algorithms, namely
for the case of Remote Radio Heads (RRH) employing both mmW and MMIMO.
-MASSIVE5Gwill propose a functional split between the data and control planes for UDNs and also defines
necessary interfaces among these functional blocks such as AP/UE dual-band discovery, mmW beamforming and
link establishment, and resource allocation and packet transmission procedures.
-MASSIVE5G will propose a converged backhaul and RRM mechanism in particular for the joint management of the
resources in the radio and BH/FH domains (optical and wireless).
-Development of an experimental testbed including FPGAs supporting the selected algorithms and optical links.
The work plan for the project is divided into four tasks. Here we just present a general overview of each task. The description of the detailed activities will be done inside on each one.
•Task 1, Definition of Scenarios Requirements and Architecture
•Task 2, Cooperative PHY Layer Algorithms
•Task 3, Resource Allocation and Management Algorithms
•Task 4, Proof of Concept
|Start Date: 01-06-2018|
|End Date: 28-02-2022|
|Team: Adão Paulo Soares da Silva, Rui Miguel Henriques Dias Morgado Dinis, Pedro Miguel Ferreira de Oliveira Pedrosa, Joumana Kassam, Atilio Manuel da Silva Gameiro, Manuel Alberto Reis de Oliveira Violas, Luís Filipe Lourenço Bernardo, Rodolfo Alexandre Duarte Oliveira, Daniel Filipe Marques Castanheira, Marco Alexandre Cravo Gomes, Hugo Alexandre de Andrade Serra|
|Groups: Mobile Networks – Av, Radio Systems – Lx, Optical Communication Systems and Networking – Av, Multimedia Signal Processing – Co|
|Local Coordinator: Adão Paulo Soares da Silva|