Open radio platforms for 5G research and beyond

The digital transformation of our society won’t be possible without the support of an appropriate mobile network infrastructure. The evolution of the mobile network infrastructure needed to meet the requirements of 5G (and beyond) poses challenging research problems: higher bandwidths and lower latencies are being increasingly demanded together with the ability to serve a multiplicity and heterogeneity of devices. On top of this, the network infrastructure evolution must take place in an economically and energetically sustainable manner.

Within this context, the existence of platforms and testbeds that can speed up mobile networks innovation and enable the validation and optimization of, e.g., new radio technologies, already in an early phase of the development cycle is of paramount importance. Such platforms and testbeds are already playing a key role in testing and evaluating the performance achieved by new research ideas and developments. Moreover, these platforms are covering a broad set of research questions, some of them focusing on specific low level aspects of wireless communications, and some others on the study of the entire end-to-end mobile network, including in some cases extensions tailored to the need of specific vertical industries (automotive, industry 4.0, eHealth, ect.). In addition, and also very importantly, these platforms and testbeds have to be able to validate the most significant network KPIs under realistic conditions and in a controllable and reproducible manner.

Thus, the main focus of this workshop will be to present the latest developments on existing platforms (both in the hardware and software domains) for 5G research and beyond, with a special focus on open source developments, as the latter facilitate innovation in the mobile networks ecosystem.

Workshop program

• Embedded SDR platform: situation and challenge (Dr. Xianjun Jiao, Ghent University)
  • In research community, most Software Defined Radio platforms are host PC based and used in office/laboratory. Researchers use desktop PC together with RF front-end to prototype new ideas quickly. However, embedded SDR platform is more suitable than PC based platform when large scale field experiment and product in special vertical market want to utilize the advantage of SDR. Embedded SDR platform, which has small form factor and can be powered by battery, are being used in-field network experiment, UAV video surveillance, mobile network analyzer and signal logger. The presentation will introduce different embedded SDR platforms, their comparison and main applications. Challenges involved in development using embedded SDR platforms will also be discussed. Finally, our thoughts and lessons learned on embedded SDR development in ORCA project will be presented.
• The Proof of the Pudding is in the Eating: Using GNU Radio in Research (Mr. Bastian Bloessl, Trinity College Dublin)
  • Software Defined Radios (SDRs), i.e., freely programmable radios, are about to revolutionize wireless. Implementing the whole communication stack in software not only adds flexibility, but also allows for rapid prototyping of novel technologies. With a proof-of-concept implementation we can advance from pure simulative performance evaluation to a combined approach with real measurements. This backs up research and speeds up development, experimentation, and testing of new concepts. The talk will provide an overview of GNU Radio, a popular real-time signal processing framework for SDRs and give ideas about how to use it for research and development.
• The Role of FPGAs and GPUs in SDR Acceleration (Prof. Ivan Seskar, WINLAB, Rutgers University)
  • In modern communication systems, flexibility at the physical layer has typically lagged since radios have mostly been implemented as fixed-function circuits, in order to minimize marginal cost, energy use, and network latency. Enabling greater flexibility would open opportunities for new wireless functions in diverse application domains. While the emerging field of Software-Defined Radio (SDR) has made significant progress toward this vision, recent results have shown that traditional SDRs suffer serious limitations mainly due to the slower sequential execution, even when using multicore CPUs. In this talk we will focus on the use of Field Programmable Gate Arrays (FGPAs) and Graphics Processing Units (GPUs) as general purpose SDR processing acceleration platforms. The talk will also introduce the RF Network-on-chip (RFNoC) open-source framework that allows easy and rapid development of FPGA processing blocks with very little overhead and with tight software integration. Finally, the presentation will conclude with the discussion on the need for a unified software abstractions in an SDR FPGA/GPU acceleration, including procedure calls, recursion, queuing, dynamic routing, shared memory and matrix algebra.
• Reconfigurable Radio for 5G: At the edge and close to the metal (Prof. Miriam Leeser, Northeastern University )
  • In the transition from 4G to 5G and beyond, adaptability and flexibility are important to support many different protocols and anticipated changes. In this research we are investigating how best to support such applications at the PHY layer using a target platform of FPGAs with embedded ARM processors. The talk will address some of the challenges at the PHY layer anticipated in 5G including digital predistortion, MIMO and beamforming as well as different tool flows to map user designs to hardware and software that make use of the latest embedded device technology and deliver high performance. A particular focus is hardware/software codesign using both FPGA resources and code running on the ARM processor.

• Extending the ns-3 LTE module for SDR: a HW-SW function split paradigm (Nikolaos Bartzoudis, CTTC)
  • In this work, we present the software defined radio (SDR) extensions and modifications of the long term evolution (LTE) module of the ns-3 simulator (LENA) to include a physical layer (PHY) implementation. The native flexibility of the simulator was combined with a hardware accelrated (HWA) PHY-layer hosted in field programmable gate array (FPGA) system-on-chip (SoC) devices. As an outcome of this work several transmission parameters can be adjusted but most importantly, the communication stack functions can be distributed across different elements of the network (denoted as function split), making this testbed suitable for network function virtualization (NFV) experimentation. Thanks to its emulated full network protocol stack, the testbed allows experimentating with versatile operating scenarios and enables the evaluation of relevant key performance indicators (KPIs) from the lower layer protocols up to application level. To this end, we present the validation of some indicative KPIs (e.g., power consumption, latency, throughput, reconfiguration time). It is important to highlight that among other benefits, function split is a technology that could help Cloud-based radio access networks (C-RAN) to tackle the fifth generation (5G) capacity and latency requirements.

• The srsLTE software radio framework (Justin Tallon, Software Radio Systems)
  • The srsLTE software radio framework
    Abstract: In the presentation we will give an overview of the srsLTE software radio framework as well as our company in general. We will briefly discuss the open source model and outline some open source grants of which srsLTE is a recipient. The technical framework discussion will detail the overall systems design, discuss some interesting optimizations as well as giving an analysis of srsLTEs compatibility with different technologies. Further, we will discuss some new features that have recently been added to the platform and give a brief overview of our future plans.

IMPORTANT DATES

Submission deadline

19 August 2018

Conference dates

18-19 September

SUBMISSION

2 page abstracts should be submitted through EAI ‘Confy‘ system, and have to comply with the Springer format (see Author’s kit section).

How do I submit a paper in Confy?

1. Go to Confy website
2. Login or sign up as new user
3. Scroll the list of conferences open for submission
4. Select CROWNCOM 2018 and navigate to the Open Radio Platforms Workshop track
5. Click the ‘submit a paper’ link and follow the instructions

Submission guidelines

Abstracts should be in English. Demonstration abstracts should be up to 2 pages in length.  The abstract submissions must follow the SPRINGER formatting guidelines (see Author’s kit section). Read the Publication Ethics and Malpractice Statement.

Publication

2 page workshop abstracts will be included in the CROWNCOM proceedings, but will not be indexed in SpringerLink Digital Library.