Last week saw the finalization of 3GPP’s next wave of 5G standards, Release 16. Despite a delay of a few months because of the pandemic crisis, the specifications have now been frozen after a series of online meetings of the 3GPP groups.
The standards body also said that the Release 16 ASN.1 and OpenAPI specifications, which usually come later than the core specs, had been frozen on schedule, and at the same time as the main standards. (The 3GPP defines ASN.1 as covering protocols between the network and the user equipment, or between network nodes, while everything else is considered a ‘functional requirement’.)
Key elements of Rel16 include many focused on the industrial and IoT applications that are often associated with 5G, but are poorly supported by the broadband-oriented Rel15. These include ultra-low latency and V2X communications. Other important projects include integrated access/backhaul and 5G in unlicensed spectrum; more efficient Massive MIMO operations; ‘six-nines’ reliability features, especially for URLLC; and others.
Potentially one of the most important elements is 5G NR-U (unlicensed), heavily sponsored by Qualcomm. 4G standards for unlicensed and shared spectrum came too late in the day to have the disruptive effect many had expected (though the USA’s CBRS band will give them a new lease of life in that market at least). But in 5G they are coming while the market is still immature, and when there is a far higher demand for a greater diversity of service providers, particularly to cater for the specific needs of different industries.
Qualcomm was instrumental in several 4G developments in unlicensed spectrum, such as the MulteFire technology, which allows LTE to be deployed without an anchor network in a licensed band, opening it up to non-MNOs. Its aim, of course, is not to undermine the operators which have made its fortune, but to expand the market for its technologies still further, at the expense of WiFi (Qualcomm has a significant WiFi business, but this has always looked like a hedge rather than the religious commitment it has to cellular).
The timing, the pattern of demand, and the capabilities of the standard all point to 5G being more successful in unlicensed spectrum than 4G was, even though it also has to find common cause with a more powerful WiFi platform, WiFi 6.
Pratik Das, a member of Qualcomm’s technical marketing department, said in a recent interview: “NR-U provides operators more spectrum options when deploying 5G, both to aggregate unlicensed spectrum with an anchor channel in licensed spectrum as well as deploying 5G NR-U Standalone in unlicensed spectrum without any licensed spectrum.”
The new standard will allow for up to 400 MHz of unlicensed spectrum to be used in the downlink direction, and up to 100 MHz on the uplink. That could see 5G competing with WiFi 6 for new swathes of unlicensed spectrum that will be opening up to support faster, cheaper and more available mobile broadband. The USA has led the way in making the 6 GHz band available, allocating a huge 1,200 MHz for unlicensed usage. The WiFi community sees this as a natural expansion band for 5 GHz WiFi, but the US regulator, FCC, was careful to mention both 5G NR-U and WiFi 6 in its statements.
Patrik Lundqvist, director of technical marketing at Qualcomm, is clear about the need to diversify the 5G service provider model, to support more business models including private industrial networks – a vision his company also had for MulteFire. He believes 5G NR-U will enable more organizations, MNOs or not, to launch private networks in areas where WiFi has, until now, been the only viable choice, especially indoors – shopping malls, enterprise campuses and so on.
Das added that 5G NR-U would complement LTE in shared bands such as CBRS. “One exciting example is to use anchored NR-U with an anchor channel in CBRS spectrum, either LTE or 5G, and aggregate it with NR-U in unlicensed spectrum, either 5 GHz or 6 GHz, to boost bandwidth and capacity,” he said.
While equipment and devices supporting Rel-16 are expected to start coming to market in early 2021, handsets with 5G NR-U support may be a bit later.
And meanwhile, despite the disruptions inflicted by Covid-19, the 3GPP is already at work on Rel-17, which should be finalized in late 2021. The key elements were agreed at the end of last year, with a package of 24 projects approved.
Rel-17 will boost overall network capacity, coverage, latency, device power efficiency and mobility, like its predecessors. A specific focus is on further advances in Massive MIMO antennas to improve spectral efficiency, and there will be work on beam management for high frequency millimeter wave bands, multi-transmission-point operations and higher mobility.
Another project will look at techniques to improve coverage for both low and mmWave bands. One way to improve coverage will be to evolve Integrated Access and Backhaul (IAB), to help enable dynamic mesh network topologies – work on that has begun in Rel-16 but further evolution is targeted.
Other highlights include:
- Mechanisms to improve device power savings.
- Mechanisms to enhance NR+LTE and NR+NR dual connectivity combinations.
- Dynamic Spectrum Sharing improvements.
- Multi-SIM will be addressed in standards for the first time.
- Multicast for public safety and venue-casting, in addition to the LTE-based 5G terrestrial broadcast standards currently being completed.
- Accurate device positioning for industrial and IoT applications.
- ‘Boundless XR’, which uses edge cloud to improve XR experiences, will be part of Release 17 work to characterize the performance of 5G networks for AR/VR/XR use cases.
- Sidelink capabilities from automotive to smartphones and public safety, including sidelink support in battery-powered devices.
- Using Release 16 sidelink for relay.
- Enhancements of Release 16 work in satellite communication, RAN slicing and self-organizing networks.
Release 17 will expand the supported spectrum range beyond 52.6 GHz, initially into the newly extended 60 GHz band, recently identified by the World Radio Conference as an IMT band for some regions. Existing mmWave waveforms used in Releases 15 and 16, for bands such as 28 GHz, can be scaled to support the higher frequencies, but this will be validated and documented in Release 17.
Enhanced support for IoT applications will be encouraged with several projects, including NR-Light, a new class of devices that will be more capable than LTE-M or NB-IoT but will support different features and lower bandwidth than 5G NR eMBB/URLLC. NR-Light will occupy just 10 or 20 MHz of bandwidth and deliver 100Mbps of downlink and 50Mbps of uplink throughput, targeting it at use cases such as wearables, and industrial IoT cameras.
In parallel, 3GPP will continue to evolve LTE-M/NB-IoT to support ‘5G Massive IOT’. That will involve improving spectral efficiency and data rates for NB-IoT, and studying how the existing NB-IoT and LTE-M platforms can be adapted, with minimal changes, to support satellite transmissions.