6G (sixth-generation wireless) will be the successor to 5G cellular technology and will use higher frequencies than 5G networks to provide substantially more capacity and much lower latency – in fact one of its goals is to support one microsecond latency communications which means it will be 1,000 times faster than one millisecond throughput.
“Society will be fully connected for the first time,” said Sarah LaSelva, Director of 6G Marketing at Keysight Technologies. “Physical and cognitive augmentation will make humans far more efficient and productive than they are today, making use of the ubiquitous nature of 6G, while new industries and business models will be enabled.
“A wide range of technologies such as AI, advanced sensors, optics, cloud computing, high-speed digital, satellite and robotics, will undergo rapid advancement in the next decade, combining and augmenting to enable new Use Models made possible by 6G.”
LaSelva concedes that there are challenges such as integrating new network topologies like global non-terrestrial networks (NTN) into existing networks, meeting user data privacy concerns, ensuring 6G is secure and creating more sustainable and green wireless systems but that the benefits will be immense.
6G will encompass many different technologies, each with its own set of technical challenges, according to LaSelva.
“One is how to train and validate the AI algorithms that will be used throughout 6G – it requires an extensive amount of data. AI tools used for image or text creation have the advantage of massive data sets available for training.
“The same cannot be said for wireless systems. Service providers often have datasets of specific wireless scenarios or the ability to capture them from their networks, but these aren’t public and privacy concerns make it unlikely these datasets will ever be public.
“Creating data sets based on a combination of real-world informed simulations and emulations is appealing but it’s an immature technology and needs to be studied and validated. After an AI algorithm has been trained, its overall performance must also be tested. AI needs to be resilient and function properly both at an algorithm level and at a system level.
“Finally, AI algorithms need to be evaluated for performance improvements – simply adding AI functionality to a wireless network simply for the sake of being able to do so won’t be beneficial.”
According to LaSelva, 6G will use multiple frequency bands so when it comes to semiconductor technology, sub 7 GHz, CMOS will continue to dominate. However, for the FR2 bands, while CMOS has acceptable levels of performance, gallium nitride (GaN) has better EIRP.
“By combining GaN with silicone (Si) to create GaN-on-Si wafers, the cost of using GaN can be reduced without sacrificing performance,” LaSelva explained.
“For frequencies in the sub-THz range, traditional CMOS consumes a large amount of power and has limitations in output power. Indium phosphide (InP) has more desirable characteristics for power amplifiers and RF components. Consequently, a mix of InP and CMOS should give a balance of high output power and efficiency for RF and compute performance for beamforming, control, and converters that will be needed.”
With new semiconductor technologies and substrates in the early research and design phase, it can be expected that innovations and creative approaches will be developed that will help to push frequencies higher.
Application benefits
In the area of communications, 6G is set to bring multi-sensory technologies that will create new ways for people to interact with each other and with their surroundings, using not just sight and sound but touch, smell, and taste as well.
“We think that 6G will eliminate physical and temporal distance between people through holographic imaging, connecting human-to-human and human-to-machine worlds,” suggested LaSelva. ”Networks will be powered by AI, allowing fully automated infrastructure optimisation and autonomous service provisioning. There will be widespread use of high-fidelity digital twins to create virtual models using past and present data and machine learning to dynamically monitor, improve, optimise, and enhance many different processes. Technologies employing extremely precise timing and orchestration of data will transform manufacturing and industrial processes as well as the networks that serve them.”
According to LaSelva, 6G will provide an environment in which all-inclusive communications are available spanning land, sea, air, and space and she expects to see commercial deployments for 6G to start sometime around the 2028-2030 timeline.
“3GPP Release 20 is likely to contain the first 6G study items and Release 21 will contain the first 6G work items,” she suggested. “Based on previous timelines for 3GPP releases, this puts the standardisation efforts starting around 2026. While it is too early to predict what will or won’t make it into the 6G standards and be deployed commercially, there are a few trends that are emerging.
“First, AI will be pervasive in wireless networks at all levels. Second, the 7-24 GHz band of spectrum is appealing, and it is likely to be used in 6G. At the same time, improvements for existing mmWave spectrum will continue. Third, the trend towards decentralised and virtualised networks with increased software functionality will continue. Finally, sub-THz spectrum is appealing for new use cases like joint communications and sensing, but it will likely not be used for traditional communications.”
With 6G being the next generation of cellular technology, it will be deployed globally but when it comes to a specific region it’s more difficult to predict.
“First deployments of 6G will likely happen in dense urban areas to reach a maximum number of potential users for the least amount of infrastructure upgrades. In the past, rural areas as well as developing nations have been slow to get upgrades to the latest G, but that may change with 6G. Non-Terrestrial Networks (NTN) have the potential to provide ubiquitous coverage satellites or other types of high-altitude vehicles instead of traditional ground-based networks.”
Deployment challenges
“As more base station and core functionality is virtualised, building new data centres will be a network deployment challenge,” said LaSelva. “Software updates are expected to be frequent in the next generation of base station technology and that will be a challenge for wireless operators, but not a new challenge in general.”
The more difficult aspect of deploying software updates will be testing and validating the new software is functional, secure, and reliable before pushing it out onto a live network, according to LaSelva.
“High fidelity digital twins will help with this as they provide an accurate digital replica of the real network where updates can be tested in a risk-free environment before going live. Network slicing is another option for testing. An isolated slice of a real network can be created to test out new features on the real-world network without interfering with the public network - both will be important tools in 6G network deployments.”
The traditional players in the ecosystem such as academic institutes, chipset providers, infrastructure vendors, carriers, and to lesser extent device manufacturers are all involved in early research, and certain governments are funding research projects and testbeds. For example, in the UK the government announced it is putting £110 million towards 6G research and development.
The trend towards disaggregated and virtualised networks driven by open RAN has brought some new players into the market, and this trend towards software defined, cloudified, virtual, and open networks will continue in 6G.
“We expect to see more non-traditional companies present in the 6G ecosystem as well as many other new entrants.”
6G aims to deliver data rates of up to 1 Tbps and to connect more devices to the network to enable massive Internet of Things (IoT), beyond the increase in connected devices we are seeing with 5G.
“With new network topologies, like integrated non-terrestrial networks, we can expect global cellular coverage, even in rural and remote areas,” said LaSelva.
AI and machine learning will be added natively to 6G networks, allowing for optimisations of every kind, at every scale like at a nation-wide level to an individual base station level.
“6G wireless networks will be more efficient than previous generations and, unlike other generations, 6G is being designed from the ground up with sustainability in mind.”
According to LaSelva, 6G will also be more secure. “6G is being designed for a post quantum computing world where traditional cybersecurity methods will no longer be sufficient to keep networks safe. New security techniques like zero-trust networks and implementing security at the physical layer are being proposed.
“And finally, not only will 6G networks be green, but by being able to connect more devices, gather more data about the physical world around us, and leverage the power of AI and ML, it will enable industries to become greener and more sustainable as well.”
