Applications of artificial intelligence in 5G and 6G networks

The artificial intelligence (AI) revolution has arrived. With the public release of applications such as ChatGPT, people are experiencing firsthand the power and potential of deep neural networks and machine learning (ML). ChatGPT is a language model that has been trained to produce texts that resemble those written by real people using massive amounts of text data from the Internet and books. This type of application is a perfect example of the advantages of artificial intelligence. It can constantly optimize the output in complex scenarios through a large amount of training data.

Wireless networks are inherently complex, generating huge amounts of data and increasing in complexity with each new generation of technology. These characteristics make AI an ideal tool for optimizing wireless networks.

The application of AI in 5G networks

As 5G technology matures, AI and ML have been introduced into research by 3GPP(Third Generation Partnership Program), an international standards organization that sets standards for cellular technology. Ai is currently being considered for air interface improvements, including network energy saving, load balancing and mobility optimization. Because there are so many potential use cases for air interfaces, only a small subset of them has been selected for study in the upcoming 3GPP R18, covering channel state information (CSI) feedback, beam management, and positioning. It is important to note that 3GPP is not developing an AI/machine learning model. Instead, it seeks to create common frameworks and evaluation methods for deploying AI/machine learning models to different functions of the air interface [1].

In addition to 3GPP and air Interface, the O-RAN alliance is exploring how AI/machine learning can be used to improve network choreography and management. For example, the architecture of the O-RAN alliance has a unique feature called RAN Intelligent Controller? (RIC)? , is mainly used to assist artificial intelligence and machine learning to optimize different usage scenarios. RIC can manage both near-real-time applications (xApps) and non-real-time applications (rApps). xApps for spectrum efficiency and energy efficiency and rApps for network orchestration and management using artificial intelligence already exist. As the O-RAN ecosystem grows and matures, more xApps/rApps and applications optimized with RIC-based AI and machine learning will emerge.

6G network native AI technology

Although 6G is in its infancy, it is certain that AI/machine learning will be a fundamental part of all aspects of future wireless communication systems. At the network level, the term "AI native" has become widely used in the industry, although there is no formal definition. One way to look at these AI native networks is to extrapolate the figure above (Figure 1) from the current virtualization and convergence trends in the RAN(Wireless Access network). Each block in the network may contain AI/machine learning models, which may vary across vendors and applications (Figure 2).

AI native networks can also be used to refer to networks built to run native AI/machine learning models. Refer to the design process below (Figure 3). In traditional 5G networks, the air interface is made up of different parts, each designed by humans. In the 5G-Advanced network, each part will utilize machine learning techniques to optimize specific functions. In a 6G network, an AI might design the entire air interface using a deep neural network.

AI/machine learning optimization

Drawing on the idea that AI/machine learning can be used to improve network orchestration and management, 6G is pinning its hopes on using AI and machine learning to solve optimization challenges. For example, AI can turn components on and off based on real-time performance to reduce power consumption across the network. Today, xApps and rApps achieve this at the base station level by turning on and off energy-intensive components such as power amplifiers that are not working.

However, AI's ability to quickly solve challenging computational problems and analyze large amounts of data opens up the possibility of optimizing network performance on a larger scale, even citywide or nationwide. The whole base station can be shut down during the period of low frequency of use, or the cell can be reconfigured to meet the real-time needs of users by using as few resources as possible in a green, low-carbon, energy-saving and environmental protection way. It is not currently possible to reconfigure base stations and entire city networks in this way, and reconfiguring and testing any changes to network configuration typically takes days or weeks. Still, the prospects for different AI technologies are so great that they remain a top consideration for infrastructure providers.

Conclusion

The use of AI in wireless networks will not wait for 6G networks. Active research is under way across the ecosystem to develop new models and integrate these into existing and future wireless communication systems. However, these models are still new and need to be evaluated for rigor and reliability. Properly training AI models on different data sets, quantifying their improvements over traditional techniques, and defining new testing methods for AI-powered modules are all critical steps that must be taken as new technologies are adopted. With the maturity of AI models, testing methods and technologies, there is no doubt that AI will revolutionize the wireless communication industry in the next 5-10 years.