News

As wireless connectivity transitions to 5G, the complexity of systems has increased, posing challenges for management. To address this, the industry is exploring new system design methods, which may involve embracing wireless artificial intelligence (AI). According to Counterpoint Research, 5G Advanced (5.5G) is expected to significantly expand the role of wireless AI across 5G networks, offering new AI applications that can enhance the design and operation of networks and devices over the next few years.
While AI has been used in smartphones and other mobile devices for some time, its use in the actual wireless network is relatively new. Traditionally, AI is employed either on the mobile device or within the network, but not both. This independent usage has limited the performance potential of end-to-end systems across devices and networks.
One reason for this limitation is that on-device AI training has only recently become possible. With on-device AI, personalized data remains local to the device rather than being shared with the cloud, improving reliability and addressing security concerns. This solution is not exclusive to smartphones and can be extended to consumer devices, sensors, and industrial equipment.
Moving forward, on-device AI will play a crucial role in the optimization of end-to-end (E2E) 5G network performance, benefiting both operators and users. It allows processing to be distributed across millions of devices, harnessing their computational power. Additionally, on-device AI facilitates personalized AI model learning based on user-specific data.
While on-device AI is a critical component in improving the E2E performance of 5G networks, Counterpoint emphasizes that independent implementation of AI does not optimize overall system performance. To achieve true E2E performance optimization, AI training must be conducted on a system-wide basis, collaboratively across both the network and devices. This requires a combination of AI expertise and deep understanding of the wireless domain.
The 3GPP’s Release 18 specification serves as a starting point for expanded wireless AI usage, expected to be employed in the future 6G networks. For example, cross-node machine learning can be utilized to adapt the Channel State Information (CSI) feedback mechanism between base stations and devices, enabling coordinated performance optimization.
Additional use cases include using machine learning to improve beam management at the base station and device, enhancing network capacity and device battery life. Machine learning can also enhance device positioning accuracy in both outdoor and indoor environments.
The implementation of wireless AI in 5G Advanced is expected to lay the foundation for even more AI innovation in 6G networks, offering various network capabilities. The native air interface of 6G AI can refine existing communication protocols and learn new protocols, enabling wireless networks to be dynamically customized to fit specific deployment scenarios, radio environments, and use cases. This will be particularly beneficial for operators as they can automatically adapt their networks to cater to a wide range of applications, including niche and vertical-specific markets.

The post Wireless AI Expected to Play a Key Role in 5G Networks appeared first on satProviders.