If you think 5G networks aren't living up to their promise, you're not alone. However, this technology is still in the early stages of implementation, although some in the mobile industry are already waiting to see what happens next. The first complete set of 5G standards came with 3GPP Release 15 in 2017, and the specifications were officially frozen in 2018. The first network deployments took place the following year, and many network operators around the world continue to deploy them to this day. Some may say that 5G has not lived up to the initial hype surrounding the technology. This was expected to not only bring about a dramatic change in download speeds, but also enable many new applications such as augmented reality (XR) and usher in a new era of connected devices. But as we wrote late last year, 5G is still relatively early in its deployment, and many networks – especially in the UK – deployed it initially by adding 5G radio antennas to the existing infrastructure. As the network core gets upgraded to support what is called 5G Standalone, 5G networks will improve. Gartner analyst Bill Menezes told us at the time: "As the carriers continue updating their networks for 5G Standalone technology, users will increasingly experience the promised improvements in speed and reliability." He added that it takes about three to five years to fully roll out a new mobile generation. 5G is, at least, delivering higher speeds, depending on location. Although the average speeds users in the UK are likely to see are lower than what it is theoretically capable of – somewhere between 75 Mbps and 240 Mbps by some estimates – this is multiple times faster than 4G. Meanwhile, the industry has not stood still, and 3GPP Release 18 marks the start of 5G-Advanced, which is expected to be frozen early in 2024 and finalized in the summer. This is forecast to bring a number of evolutionary improvements as well as new features, that may address the limitations of the initial 5G rollouts. 5G-Advanced introduces support for non-terrestrial networks (such as satellite connections) increasing coverage in remote and rural areas, and is claimed to offer better uplink speeds. While mobile broadband and Internet of Things (IoT) were held up as 5G use cases, 5G-Advanced promises to hike mobile broadband performance, support for new applications, and boost intelligent network automation, at least according to a whitepaper commissioned by telecoms outfit Ericsson. Extra performance comes from MIMO (multiple input, multiple output) signal capacity being boosted in both uplink and downlink with support for enhanced demodulation reference symbols. The new applications being touted are extended reality (again) for uses including remote control and industrial automation, plus indoor positioning and IoT. According to Ericsson, performance for extended reality services comes from support in the 5G network core for data rate adaption using the low latency low loss system (L4S). Intelligent network automation is, we're told, supposed to make use of artificial intelligence and machine learning techniques to improve network performance, in a similar vein to AIOps in datacenters. Advances in the 5G network core architecture for analytics and data collection should serve as a good foundation for AI/ML-based techniques, according to the whitepaper. Chinese technology giant Huawei claimed in June that next year it will launch all the kit a network operator needs to run a 5G-Advanced network, although for some reason it is labeling the technology as 5.5G, to the irritation of a few in the industry. But perhaps 5.5G isn't such a bad name: Ericsson regards 5G-Advanced as a stepping stone towards future 6G network standards. That next generation tech is a long way off, with specifications not likely to be settled upon until sometime around 2028, and initial deployments tentatively pencilled in for 2030. However, the 3GPP standards body – actually an umbrella group for seven global telecommunications standards organizations – is likely to commence early work on 6G in 2024. So what can we expect to see from 6G networks? More of the same, it would appear at first glance – higher speeds, more pervasive networks, more support for demanding applications such as extended reality. So do we really need the industry to deliver yet another network standard? According to Peter Vetter, president of Bell Labs Core Research at Nokia, we certainly do, if only for the greater energy efficiency that the industry is aspiring to deliver for 6G networks. "This is an important research question because we can see mobile traffic going up over the next decade by a factor of 10 or even a factor of 20. So if we don't do anything, the power consumption of the base station will go up by a factor of 10 or 20," Vetter told The Register. The goal is to halve the total energy consumption of mobile networks with 6G, which means that energy efficiency will have to improve by a factor of 40, if network traffic goes up by a factor of 20. "This requires some foundational research, and even if there are people that say we don't need 6G, we are saying 'yes, you do need 6G' because traffic analysis shows that 5G is going to run out of steam by the end of the decade," Vetter said. This, he emphasized, means 5G networks won't have the capacity to cope with the increase in traffic. Nokia is advocating for 6G spectrum just above the current mid band range for 5G (1-6 GHz) because that will allow for deployment from existing cell sites, he added, and this will call for large-scale antenna arrays that can better direct the electromagnetic energy. (A key topic of discussion at the recent ITU World Radiocommunication Conference, WRC-23, was said to be utilization of the 7-15 GHz band as the main spectrum for 6G.) "We need new concepts like hybrid beam forming, new algorithms and sleep modes and antenna muting," Vetter said, "so there are all kinds of concepts to improve the energy efficiency in absolute numbers." According to Telefonica, 5G networks are already up to 90 percent more energy efficient per unit of traffic than older networks, with up to 70 percent of all the network power consumption in the radio access network (RAN). One idea for energy efficiency is the use of AI to manage the infrastructure, as is being touted in 5G-Advanced networks. This will not just look after the core network, but also the wireless air interface at the base stations where it can be used to learn the channel conditions and how to adjust configuration for optimal performance. Vetter claims that Nokia has already tested a proof of concept for this, which shows that networks can get 30 percent better capacity over the same radio channel using AI in the base stations. Nokia is the lead developer on 6G-ANNA, a "lighthouse project" funded by the German government aimed at driving global pre-standardization activities for 6G. Other companies taking part include Vodafone, Siemens, Ericsson, and Bosch. Early concepts for 6G under this project foresee peak data rates exceeding 100 Gbps, depending on the spectrum used, but this may vary depending on the spectrum available in different regions. Other concepts for future 6G networks include network sensing and even greater connectivity for industrial automation and IoT projects. Network sensing involves the network using its own radio signals to monitor the environment, Vetter said. "All those radio systems around us, they can also be used as a sensor. If you're smart about it, you can use existing base stations to monitor people and things to optimize movement in an airport or a city environment," he said. Nokia has built a proof of concept, using an existing 5G base station hacked to operate as a radar, and researchers were able localize people and detect movement within an accuracy of less than a meter, Vetter claimed. On the connectivity side, Nokia believes wireless networks will eventually replace wired connectivity, even in enterprise and industrial sectors, which may place greater demands on them to support all those devices. For this reason, the plans are for 6G to support 10 times more connected devices than 5G, Vetter said. Overall, the concepts for 6G so far seem to center on mobile networks becoming more pervasive and creating capacity and performance for demanding applications like telepresence, as well as connecting myriad sensors and devices beyond phones. But all of this costs money in terms of R&D dollars, in addition to building and deploying the infrastructure. A recent article in the Financial Times estimates that spending by network operators is set to slow next year, and that operators want to see better returns on their existing investments in 5G before they consider further expensive network upgrades. And here comes the rub: That position may prove untenable as operators feel pressure to keep investing for fear of falling behind rivals - the same dynamics as in previous rollouts in the mobile network industry.