Keysight Technologies is leveraging the power of artificial intelligence (AI) and machine learning (ML) to solve some of the most pressing challenges in 5G and emerging 6G networks, a bold move toward redefining the future of wireless communication.
Wireless networks have grown exponentially more complex with each generation, producing massive volumes of data. Keysight recognizes that this combination of complexity and data abundance makes wireless networks an ideal candidate for AI-powered optimization. The company is now applying mature AI technologies, such as deep neural networks and machine learning, to optimize these intricate systems, similar to how AI has transformed speech transcription and natural language processing in consumer applications like Descript AI and ChatGPT.
In the 5G space, Keysight is actively engaged in aligning with the 3rd Generation Partnership Project (3GPP), the international standards body responsible for cellular technologies. While 3GPP does not create AI/ML models, it focuses on establishing frameworks and evaluation methods for models applied to air interface functions. Keysight is contributing to studies under 3GPP Release 18, targeting use cases such as channel state information (CSI) feedback, beam management, positioning, energy savings, load balancing, and mobility optimization.
Beyond 3GPP, Keysight supports the O-RAN ALLIANCE’s vision for intelligent and open networks. This includes the integration of AI-driven applications within the RAN Intelligent Controller (RIC). Keysight is helping to shape and test xApps, which run in near real-time, and rApps, which operate in non-real-time. These applications already demonstrate potential in improving spectral efficiency, energy usage, and orchestration across the radio access network.
Looking ahead to 6G, Keysight anticipates a future where AI is not just an optimization tool but a foundational element of wireless network architecture. The term “AI native” is gaining traction, referring to networks that are inherently built to run AI/ML models across every component. In this vision, AI could be responsible for designing entire air interfaces using deep learning, with network components embedding ML models tailored by vendors for specific use cases.
Figure 1: ORAN Network
One way to look at these AI-native networks is to extrapolate the diagram above (Figure 1) based on current trends of virtualization and the RAN (Radio Access Network) disaggregation. Each network block will likely contain AI/ML models varying from vendor to vendor and application to application (Figure 2).
Figure 2: ORAN 6G Network
AI-native networks can also mean networks built to run AI/ML models natively. Consider the design flow below (Figure 3). In traditional 5G networks, the air interface comprises different processing blocks designed by humans. In 5G Advanced, each block will leverage ML to optimize a specific function. In 6G, AI may design the entire air interface using deep neural networks.
Figure 3: Progression from AI-infused to AI-native networks
One of the most promising applications of AI lies in large-scale network optimization. While current technologies like xApps and rApps can manage power consumption at the base station level, Keysight envisions a future where AI enables real-time, city- or even nation-wide network reconfiguration. This could mean entire base stations shutting down during periods of low demand and reactivating dynamically, all in an energy-efficient manner.
However, Keysight emphasizes that integrating AI into wireless systems requires rigorous testing. The company is prioritizing the training of AI models on diverse datasets, benchmarking their performance against traditional techniques, and developing new test methodologies. As these practices mature, Keysight expects AI to play a revolutionary role in shaping wireless communication technologies over the next decade.
By staying at the forefront of AI integration, Keysight Technologies is not only addressing current challenges in 5G but also laying the groundwork for AI-native 6G networks. The future of wireless communications, according to Keysight, will be intelligent, adaptive, and driven by data.
Click here to learn more about Keysight Technologies.
