Right now, at the dawn of the global rollout of 4G networks, the term 5G has started to appear, although it has yet to be defined. Opinions are divided on whether the term 5G will ever be standardized.
Within a decade, 5G networks could be on the way, at least if you look at the historical evolution of mobile networks with the launch of NMT (a Nordic 1G network) in 1981, GSM in 1992, 3G in 2003 and 4G in 2009.
Kick-starting the 5G-era
In January 2011, the International Telecommunication Union (ITU) approved the IMT-Advanced specification. ITU Secretary-General Hamadoun TourĂ© said that kick-started the 5G discussions: “Compared to IMT-Advanced, today’s smartphone will feel like an old dial-up Internet connection. Internet access, streaming video and data transfer will be possible anywhere and anytime and with a higher capacity than most wired PCs today”.
However, the new standard is not 5G but rather an update within the 4G framework.
The Long Term Evolution: 4G
In 2012, 4G based on LTE is expected to be the prevailing technology for new mobile networks. Operators currently operating 3G networks will be forced to upgrade to 4G within a few years to handle the increasing amount of traffic on the networks.
By 2020, it is reasonable to imagine that LTE will be the prevailing standard globally and that 3G networks will be retired or are already retired.
LTE stands for Long Term Evolution, intending to be a mobile network that evolves over time. Therefore, any transition to 5G will need to start from 4G and consist of an update of the existing networks.
Another factor in favour is the availability of frequencies: the ITU is the UN body that administers the International Radio Regulations. Its turnaround time from the idea to rolling out the technology to users is 10-12 years. Setting aside new frequency bands for 5G doesn’t make sense, so the next generation of mobile networks will likely share frequencies with current mobile networks.
New network, new base-station?
An additional factor in favour of a smooth transition is the OFDMA modulation technology, which will likely remain the dominant technology in the future. The move from 4G to 5G will not be a total restructuring of the infrastructure as it has been from 1G via 2G and 3G to 4G. Base station manufacturers have driven generational evolution. A new generation – a new “G” – has meant new technologies, higher speeds and, by default, new sales of base stations and terminals.
Operators have been mainly served a ready-made set of requirements, but increasingly they now want to make their voices heard. The Next Generation Mobile Network (NGMN) operator initiative aims to develop the next generation of mobile networks. Japanese operator NTT Docomo is one of the most active NGMN members in the 5G discussion. The operator is working on something it calls NMN, the Next Mobile Network, which according to NTT Docomo, is the last step before 5G. According to NTT Docomo, NMN could be standardized in 2014 and launched in 2018, followed by 5G, which will begin development in 2018 and launch in 2020.
The company’s Networking Research Group (NRG) is responsible for researching the technologies it believes will form the architecture of future mobile networks. The systems should be able to deliver high connection speeds and availability wherever the user needs it at that moment. The networks should be scalable, cost-effective mobile network platforms that can efficiently and dynamically load-balance network capacity as demand for services changes over the day.
One of NTT Docomo’s visions for the NMN is to deliver the capacity required to ensure that the user does not feel the limitations of the network – that the perceived speed is the same regardless of technology or geographical location. NTT Docomo’s version of the future mobile network is reasonably consistent with developments in the market at large. However, whether or not progress should be labelled 5G is an open question.
To cope with the future surge in mobile network traffic, networks need to be densified compared to current networks. They also need to be combined as there is a general need for more spectrum. Work on network architecture will be more crucial than developing radio interfaces.
The need for capacity
Regardless of when or if 5G is launched, the need for capacity is unlikely to slow down. Developments and new applications will require increased speed in mobile networks. One estimate is that we can expect a doubling of capacity needed each year.
How do we manage capacity demand?
The number of base stations in the 4G network can only be increased for a while. Instead, a likely evolution is that a new base station model will be used indoors and another model outdoors to cope with capacity needs. These so-called HETNETs are mixed networks based on several different wireless technologies, with traditional base stations based on one technology, such as LTE, and so-called pico base stations based on another standard.
The first tentative step towards a pico base station standard is the 802.11u network standard, which will be able to use wireless networks for handover from mobile networks. 802.11u is likely to be found in products launched on the market as early as this year.
Pico-base stations are a potentially huge market, with the number of pico-base stations likely to significantly exceed the number of regular base stations. Pico base stations can connect directly to fibre or cat5 networks and establish local mobile networks for homes, offices and shopping malls. Thanks to these local indoor base stations, the rest of the mobile networks – the outdoor networks – can be effectively relieved.
What spectrum to use?
An open question is which frequency spectrum to use for pico-base stations. 5.6 GHz is one way. Another is secondary spectrum use. With white space technology, parts of the spectrum can be reused indoors. This would be especially beneficial for short-range transmission. Whitespace is based on so-called cognitive radio with dynamic spectrum access that decides for itself which frequency band can be used. The advantage is that the technology consumes less of the naturally available frequency spectrum. One of the disadvantages is that it needs to be known in advance how much spectrum is available for each location. Other frequency bands, such as 60 GHz, are reserved for future short-range broadcasting but would be unsuitable for indoor mobile networks.
HETNET is essential for future mobile systems. One feature we will see more and more is that networks are becoming heterogeneous – the best communication link at the moment is used regardless of technology. The bottom line is that HETNET will reshape the mobile industry significantly. Issues such as who builds the infrastructure, what company produces the base stations indoors, and how operators will charge the home base station operator are intricate. If one were to define 5G based on the impact of the industry’s business models and how we define mobile networks, then HETNET is likely to be the definition of 5G.
If indoor networks are based on pico-base stations, outdoor base stations in future mobile networks will rely on simplified operations, more accessible network management and higher energy efficiency. The cost of each packet transmitted must be reduced. For LTE-Advanced, CO-MIMO will be introduced, a technology that will make base stations interact to a greater extent than before. CO-MIMO, a cooperative MIMO, is based on splitting the signal between multiple cells – useful when a terminal is located in a sector covered by two base stations. Signals can be shared between base stations to maximize capacity, distribute load across the network and reduce interference.
Improved MIMO
Technologies such as CO-MIMO are considered to be very important in next-generation mobile systems. These technologies place high demands on synchronization between base stations and a well-provisioned underlying infrastructure. The main obstacle to universal deployment is that it requires dedicated core networks – IP-based networks are too slow to support the technology. Future mobile networks are likely to rely partly on technologies such as CO-MIMO in areas with high capacity requirements. In contrast, other network parts depend on a distributed infrastructure over IP to squeeze costs and increase capacity.
Adaptive antennas
Adaptive antennas are a critical component to meeting the increased bandwidth requirements. Adaptive antennas are used to improve the spectral efficiency of networks – in other words, to increase data rates without simultaneously growing bandwidth requirements. Adaptive antennas in MIMO systems include terminals with adaptive antennas but with fewer antenna elements than the base stations. An exciting avenue for future networks is distributed antenna systems, which spread many antenna elements over an area and then link the signals from each antenna unit to a centralized processing unit. However, it is debatable whether this technology deserves to be called 5G or is just a further development of 4G. Another exciting avenue is communication using large antenna arrays with hundreds of antenna elements on the base station side and simple antennas on the terminal side. Theoretically, such systems can offer network capacities on par with larger MIMO systems.
The allocation of capacity in mobile networks is also crucial in the future mobile network – it is about maximizing the perceived quality of service and allocating the right resources to the right place for each connection. One step along the way could be predictive caching, a technique to predict what information users will request and store in advance. The method is already being tested and allows terminals to store information during times when network stress is low, such as at night or when they are connected to wireless networks. By holding more information in the terminals, less data can be retrieved when the load is high; tests show that as much as 30% of capacity could be saved with this technology.
What is 5G?
An attempt to define 5G could be “intelligent wireless technology that can connect the world without boundaries”.
The question is whether the term 5G is overshadowed by the exciting technology developments ahead. In the future, 4G or 5G will likely change how we connect to the outside world. We will move from mobile networks being an alternative means of connecting to the outside world (4G) to a primary means of connecting to the outside world (5G). The pico-base at home may well replace our need for wifi, and we may well come to regard the mobile network as our primary and only wireless connection path. Future technological generational shifts in mobile networks are unlikely to cause the same stir today. Thus, we will define generations differently than in the past. The operators will focus on creating more capacity in the network rather than labelling their technology. The naming of mobile networks as “4G” or “5G” is perhaps not as important an issue.