5G: looking for the next generation wireless infrastructure

Feature articles |
By eeNews Europe

What we have today are a collection of research projects looking at what 5G could entail, ranging from massive MIMO to the use of mmWave frequencies. One of the risks to 5G is that LTE-A adopts some of these techniques, much as 3G boosted performance as 4G was being defined. It is certain that 5G will have to use higher frequencies than today to get the required bandwidth necessary to support such high data rates. Further cells will need to shrink significantly, to use spectrum as efficiently as possible.

According to John Spindler, Director of Product Management at TE Connectivity 5G could be ready to start implementation around 2020, but for this to happen specifications would need to defined in around a year or two from now.

Recently, TE Connectivity announced that its FlexWave Prism and FlexWave Spectrum distributed antenna systems (DAS) had been deployed for the 49th NFL championship game that took place on February 1 in Phoenix, Arizona at the University of Phoenix Stadium. The massive deployment includes 96 main hubs, 49 expansion hubs, and 225 remote antenna units to cover the stadium bowl, luxury boxes and service areas. The system supports various 700, 800, 850, 1900 and 2100 MHz LTE, CDMA, EVDO and UMTS services.

John Spindler says that going from 3G to 4G with DAS has not posed any problems and expects the same for 5G. However, before any real discussion can take place the operating frequencies for 5G need to be decided and the air interface protocol needs to be defined. Essentially DAS is ready for 5G, the radios will need to be upgraded, but most of the infrastructure in place can be reused. This is in contrast to small cells, which will struggle to cater to 5G without a hardware upgrade.

One area where 5G is beginning to take shape is in test and measurement. National Instruments are involved in a number of projects looking at 5G. 5G covers a multitude of emerging technologies along with extreme bandwidth and data rate requirements. These include advances in HetNet architectures, air interfaces that deploy 3D MIMO or massive MIMO, new modulation schemes and possibly the use of higher frequencies such as mmWave. Some possible modulation schemes for 5G include GFDM, Generalised Frequency Division Multiplexing, FBMC, Filter Bank Multi-Carrier, and UFMC, Universal Filtered Multi Carrier amongst others.

The point being that there is a lot of potential technology that could eventually make up the outlines of a 5G standard. However, to get there, researchers need to be able to test their prototypes. National Instruments is heavily involved in providing test systems so researchers can explore emerging wireless technology.

For example, the company is Nokia to collaborate on advanced research related to 5G, such as exploring peak data rates and cell-edge rates in excess of 10 Gbps and 100 Mbps, respectively. Nokia plans to demonstrate the viability of high-frequency millimeter wave as an option for 5G-radio-access. An experimental 5G proof-of-concept system will be implemented using LabVIEW and PXI baseband modules from National Instruments — to provide a state-of-art experimental system for rapid prototyping of the 5G-air-interface.

Separately, the PXI Express platform based on LabVIEW has been configured to perform all the signal processing, synchronization, control functionality, and I/O necessary to implement the wireless protocols required to meet 5G requirements. When configured appropriately, the modular nature of these platforms provides the flexibility needed to achieve the 10 Gbps, per user target data rate for 5G cellular access technology, and orders of magnitude higher for mmWave backhaul needs. A large-scale European project researching wireless communications, miWaves, uses a similar setup. The miWaveS project is focused on the V-band (57‒66 GHz) and the E band (71‒76 GHz, 81‒86 GHz). The project started in January 2014 and will terminate in December 2016.

Also active in 5G is Anite, which is involved in the EU funded project METIS. This project aims to lay the foundation for 5G. Anite is involved in leading channel model research in this area.

The Anite-led task group within the METIS project recently published the first channel models for 5G. An essential step towards further development of candidate 5G technologies, the interim channel models were co-authored by eight METIS partners and approved by other key members for publication.

The technical requirements for 5G will be very challenging, thus testing the radio channel is even more important compared to 4G or 3G. It is expected that 5G will adapt to various radio channel conditions in a more efficient way, utilising all dimensions of the radio channel such as delay, frequency, time, location, elevation and polarization.

James Goodwin, Director of Product Management at Anite expects that more will be squeezed out of the LTE standard by improving the air interface with the use of more complex MIMO and even massive MIMO, as well as by allocating new spectrum.

James Goodwin contends that 5G will only be addressed properly at the mmWave level with new modulation schemes. In order for this to progress it is essential that the real-world channel between the radio and mobile device are well understood. As part of the METIS project Anite is building tools to evaluate such channels with different modulation schemes and is leading channel model research in this area

For 5G to go global one key issue is the need for a unified spectrum. James Goodwin adds that with 5G it is very difficult to do a multi-frequency mobile device at 30 GHz. A lot of people are skeptical whether mmWave will be a practical answer, but mmWave is beginning to appear in Wi-Fi products and costs are expected to drop, while R&D accelerates.

Anite is also a member of Project Virtuoso, an Intel-led industry project that is researching testing environments to accelerate 4G and 5G technology development and testing. Within this project, Anite aims to enhance its Virtual Drive Testing Tools (VDT) to utilise data measured in the field to “virtually” recreate the field test environment in a laboratory.

According to Janne Kolu, Director, Channel Emulator Products at Anite, by using a variety of tools to take in-field measurements, the data collected can then be used to simulate the environment in the lab. By replaying this data in the lab it is currently possible to accurately simulate power level requirements. The next steps are to implement fast forwarding, multipath and spatial radio channel requirements. By the time the project ends in 2018 some 5G topics will have been addressed.

To conclude, for 5G to start happening a frequency plan needs to be specified followed by a definition of the air-interface. A modulation scheme will need to be adopted, as has been the case for every wireless generation. In general, 5G will need to deliver a peak data rate of round 10 Gbps with a cell edge rate of around 100 Mbps, along with latency of under 1-ms to warrant being the next generation wireless standard. This might seem a tall order, but 4G in the form of LTE and LTE-A will continue to evolve over the next few years.

Related articles:
Gain early access to 5G capabilities
5G research and trial investments to hit $5 billion by 2020
Agilent and China Mobile to collaborate on 5G wireless
National Instruments joins CROWD for 5G wireless networks research
Tiny, cheaper radio wave circulator could double the useful bandwidth
The battle for the future and viability of 5G is in R&D


Linked Articles
eeNews Wireless