The past 2019 can be called the "year of 5G". In April, the 3GPP consortium, which develops the mobile specifications, released its 15th next-generation standards release, and networks began to roll out around the world. Clarification of 5G parameters is still ongoing, and releases 16 and 17 should appear in 2020-2021, which will complete the description of 5G, bringing it to the conditional level of "5 ++". Meanwhile, the race to the next generation 6G has already begun.
In March 2019, the first meeting of the 6G Flagship consortium took place at the Finnish University of Oulu. The university, which is Nokia's key R&D base, has spearheaded work on next generation networks. And in November, the government of China officially launched the development of 6G technologies. All major telecommunications equipment manufacturers have already joined them, and the next 6G Flagship meeting is due to take place in March 2020.
“The issue of 5G can be considered generally closed at release 15,” Vitaly Shub, head of Skoltech's leading research center, told us, who is directly involved in the work on the new generation of communications. - The specifications have been determined, the technologies have been created, the industrial production of equipment is underway. Chinese factories produce about one hundred thousand base stations per month. It's time to think about what the 6G connection will look like.
Eternal cycle
Telecommunication infrastructure uses two fundamentally different types of networks. Fixed-resource networks - such as, for example, a wired connection over copper, coaxial or fiber-optic cable - directly connect a subscriber to a port of the operator, which guarantees a certain bandwidth of this channel. A dedicated connection is intended for the user personally, like a water pipe connected to a tap in a house.
In contrast, cellular networks are by definition divisible networks. Their specification guarantees a certain transfer rate to and from the general pool of subscribers only between them and the base station. However, the final data exchange rate depends on the number of connected subscribers, network capacity and other factors. “In fact, mobile communication up to the 4th generation inclusive is a unique business that can provide a service without any guarantees of its quality,” says Vitaly Shub. “Moreover, there is nothing to be done about it: such a feature follows from the very“physics”of the network, from the limited resources of its resource, which is shared among all users."
Promotional video:
As a result, each next generation of cellular communication goes through the same characteristic stages. The first time after the appearance of the new technology, there are not too many subscribers in such a network and the speeds available to them are really high. However, then the network begins to fill up, and there are more and more users and demanding applications. As a result, speeds drop and there is a need to introduce new technologies and a new generation of communication. Practice shows that such a change takes about 10-12 years.
“Business develops along the saw line: the gradual saturation of networks ends with the emergence of the next generation of communication, which reduces this load,” explains Vitaly Shub. - First, there is a supply, it creates a demand for new opportunities. But then everything changes: the emerging demand requires a new supply, new technologies to satisfy it. Cellular operators are simply forced to constantly expand the network and improve its characteristics."
Between the fifth and sixth
Each next generation of cellular communication can be associated with transitions to new, more and more complex principles of signal coding. The first of these used frequency division multiplexing (FDMA) systems, the simplest approach in which access to a common channel is divided between users by temporarily assigning specific frequencies to them. Next, TDMA technologies became widespread, allowing several subscribers to use the same channel, sharing it in short time intervals.
Then, code division multiple access (CDMA and WCDMA) was introduced, which provides additional opportunities for parallel use of frequencies. In this case, the signal is modulated with a special coding sequence, for each subscriber its own. The base station antenna transmits an entangled, noise-like signal, but each final recipient, knowing his code, is able to extract the part he needs from it.
Orthogonal carrier multiple access (OFDMA) was then implemented, in which each carrier frequency, in turn, is divided into multiple subcarriers modulated independently of each other. Today this approach is approaching its theoretical limit. “For each technology there is a limiting spectral efficiency, that is, the number of bits per second that 1 Hz of radio waves can transmit,” explains Vitaly Shub. - The fifth generation is approaching 30-50 bit / sHz, almost completely using the capabilities of the mathematical coding apparatus. This gives huge bandwidth: add ultra-wide carrier bandwidth and you get numbers from 100 Mbps to 1 Gbps, and in some cases even 20 Gbps."
It is expected that 6G communication will reach already from 100 Gbps to 1 Tbps, and the network response speed - less than a millisecond. The exact requirements for the standard have not yet been formulated, but it is assumed that these are the numbers that will be needed for the operation of unmanned vehicles, complex artificial intelligence and virtual reality systems, robotic industry and logistics. Achieving the desired indicators will require the use of new frequencies, new mathematics and even physics.
New speeds
Data rate is determined by the bandwidth and spectral efficiency, and work for 6G is being done in both directions. So, in order to increase the carrier width, it is necessary to use a new range that is not yet available for communication, moving to even shorter-wave radio waves - with a frequency of up to 100 GHz and even higher, in the terahertz, submillimeter region (300 GHz - 3 THz), which remains practically unoccupied and allows you to use a wide working range.
Until recently, terahertz transmitters and receivers remained as complex and cumbersome as early computers. Such installations have found widespread use only in recent years - for example, during the examination of baggage in search of explosives, in medicine and materials science. For the sixth generation of communications, terahertz devices should become even more miniature and energy efficient. And in addition to this wide channel, new signal coding technologies should appear to increase its spectral efficiency. One of the key areas of this work has become "optical vortices", which are actively pursued by developers from Skolkovo. “A light wave can be imagined as a corkscrew or a spiral,” explains Vitaly Shub. - The pitch of this spiral can be uneven, moreover, it can be controlled. Having learned to modulate such wave irregularities,we get an additional way to encode the signal. " Such technologies are moving forward by leaps and bounds, and in 2018 Australian scientists scaled down the system for modulating the angular orbital momentum (OAM) to the size of a microchip, quite suitable for use in a pocket gadget. According to some estimates, the use of OAM coding will increase the spectral efficiency by at least five times. “The theoretical limits have not yet been established here, since it is not yet clear how much we will be able to vary and control the“beam step,”adds Vitaly Shub. "It is possible that the growth will be ten or a hundred times."and in 2018, Australian scientists scaled down a system for modulating the orbital angular momentum (OAM) to the size of a microchip, quite suitable for use in a pocket gadget. According to some estimates, the use of OAM coding will increase the spectral efficiency by at least five times. “The theoretical limits have not yet been established here, since it is not yet clear how much we will be able to vary and control the“beam step,”adds Vitaly Shub. "It is possible that the growth will be ten or a hundred times."and in 2018, Australian scientists scaled down a system for modulating orbital angular momentum (OAM) to the size of a microchip, quite suitable for use in a pocket gadget. According to some estimates, the use of OAM coding will increase the spectral efficiency by at least five times. “The theoretical limits have not yet been established here, since it is not yet clear how much we will be able to vary and control the“beam step,”adds Vitaly Shub. "It is possible that the growth will be ten or a hundred times."It is possible that the growth will be ten or a hundred times."It is possible that the growth will be ten or a hundred times."
Record reactions
The need to bring the response time of 6G networks to sub-millisecond levels poses completely different problems. According to Vitaly Shub, this will require global changes in the network topology. The fact is that in recent years they have developed with a focus on "cloud" data storage. Our files, music, photos can be physically located anywhere, on a server in the USA, Australia or Denmark. As long as the "bottleneck" in access to them is the wireless speed, this does not really matter. However, 5G communication is already fast enough, and even the most powerful wired channel between the cellular operator and the server is not enough: storage must be moved closer to the subscriber. “Everything is starting to return to normal,” says Vitaly Shub. "What moved in one direction in the third and fourth generations turns back."This approach embodies the concept of Mobile Edge Computing (MEC): packet switching centers that accumulate the data most demanded by users to speed up access to them move as close to the recipient as possible, and smart software constantly adjusts the content and distribution of content depending on the needs of the subscriber … Instead of a high, multi-tiered hierarchy, the network becomes almost “flat,” and the latency within it drops dramatically.multi-tiered hierarchy, the network becomes almost "flat", and the latency time within it is sharply reduced.multi-tiered hierarchy, the network becomes almost "flat", and the latency time within it is sharply reduced.
The MEC implementation faces a number of new and unresolved technical challenges. In particular, an even greater miniaturization of signal packet switching systems and data storage devices is required, an increase in their capacity and a decrease in power consumption. In the meantime, 6G is making only the first rough steps in anticipation of the time when the previous generation will approach the "saturation stage". Most likely, this will happen around 2025-2027, when new subscriber and application requests become clear. Only then will the specific requirements for the following communication standards be formulated.
Political generation
The main players in this field have already been identified - apart from Nokia and the Chinese Huawei, these are Samsung and Ericsson corporations. It is expected that around 2028-2030 they will develop the basic parameters of 6G, and the 3GPP consortium will release another release describing the key standards of the next generation. However, everything is capable of going according to another, unexpected scenario. “One can expect that the sixth generation will become the most politicized,” says Vitaly Shub. "Attempts by the West to" curb "China are already apparent at the 5G stage, and they can continue, destroying the entire complex system of international cooperation." Indeed, China's Huawei owns almost a third of the patent pool for 5G communications technology - and with the sixth generation, this situation is likely to only get worse. In addition to the already adopted state program for the development of 6G,The PRC can rely on internal resources inaccessible anywhere else in the world, on its huge market and colossal volumes of "big data". “The entire modern economy is a livestock economy,” adds Vitaly Shub.
However, within the framework of such an economy, Russia still retains its own small unique niche. Our developers are actively involved in creating the physical and technological foundation from which both patents and 3GPP standards will emerge. “These are new materials, new mathematics, new principles - a nightmarish work in terms of volume,” sums up Vitaly Shub. “We can only hope that we will be able to meet the usual 10-year implementation cycle.”
Roman Fishman