Leveraging technological advances to drive more sustainable, high-performing communications networks is quickly emerging as a key consideration for modern CTOs. For me, the potential of artificial intelligence (AI) and O-RAN presents the most compelling opportunities.
The mobile communications business is ambitious, with more than 60% of the division’s revenues committed to science-based targets to rapidly reduce carbon emissions over the next decade. Clearly, this will require investments in new technologies, such as AI, that more effectively leverage the full capabilities of the underlying technology to reduce scope 3 emissions.
At the same time, the world has not stood still. Data consumption continues to grow relentlessly, driven by a set of new applications that promise a higher quality customer experience and new industries looking to leverage the capabilities of wide area network connectivity to reduce their emissions. increase. So the key question we have to ask ourselves is how does the telecom industry reduce its carbon footprint while handling the ever-increasing volume and variety of data?
O-RAN and AI – Use Cases for AI in Communications to Drive Energy Efficiency and Automation
Fortunately, there are technology trends that are ushering in new innovation streams that support our net-zero efforts. At the center of this is O-RAN. Its open interfaces allow different vendors to innovate in specific niches across network solutions, creating a thriving market at the subsystem level.
At the same time, more telcos are looking to extract more value with O-RAN interfaces that use machine learning to analyze published data. A Cambridge Consultants 5G testbed study (detailed in the O-RAN white paper) shows that advanced algorithms enable greater automation to maximize service performance across the network and improve service user engagement. I knew I had the opportunity to transform. experience.
O-RAN and AI have the potential to enable more efficient use of infrastructure, reduce network costs, and drive more efficient fault finding and network operations by increasing the level of network automation. There is also
This means that far fewer network resources can be used to achieve a higher quality of experience, breaking the established pattern in the mobile industry of simply over-provisioning capacity to achieve the desired quality of experience. can do. This is a proven strategy so far. Resources of all kinds have been wasted, furthering the commoditization of network infrastructure.
Our Findings – Transforming User Experience
The benefits of O-RAN and advanced machine learning extend to the point where the network can be directly optimized for individual user/application performance.
Our research showed that by analyzing the data published by the network, we can infer real user experience of a given application without affecting user privacy. Therefore, understanding the user experience, especially when combined with other data sources, will enable you to create unique service offerings from standardized network infrastructure.
The advantage of this is clear. Maximize the quality of experience on an individual and collective level without compromising privacy. We maximize network utilization and efficiency while minimizing costs and environmental impact. And build a platform for innovation and revenue growth through immersive media.
Achieving these goals requires expertise across different competence areas. You need to understand cellular network architectures and how they are designed to achieve the most efficient coverage and experience for your particular investment.
It also requires a deep understanding of network data available through O-RAN interfaces and what this data reveals about service and network conditions. Finally, it also requires a deep understanding of the nature of machine learning algorithms and how to develop the most applicable algorithms to achieve desired results.
Pictured above: Based on a video conferencing application, Cambridge Consultants’ algorithm (yellow) accurately predicts a user experience that matches the actual user experience (purple). Quoted from “O-RAN and AI: Optimized Networks Ready to Transform Service Performance,” Cambridge Consultants, 2023
Evolution to virtualized software-based implementations
The underlying technologies and architectures of network infrastructure are undergoing fundamental changes. We are rapidly evolving from bespoke, hardware-centric implementations to virtualized, software-based implementations, to AI-native technologies where many aspects of our technology stack are based on non-deterministic database technologies. I hope. Our recent white paper, O-RAN and AI: Optimized Networks Are Ready to Transform Service Performance, explores this point further and shows how Cambridge Consultants use AI to We present a completed study to demonstrate the potential of this approach by inferring experiences. It describes a system designed to estimate real-time user experience and provides guidance on how to deploy other network features to maximize that experience.
