INNOVATIONS UNPLUGGED – the Rohde & Schwarz technology podcast · Rohde & Schwarz

[MWC2026 special] No dropped calls, no drama: AI reliability boost for your mobile connectivity

·21 min

The episode opens by contrasting the louder headlines around AI that writes, speaks, draws, and codes with a quieter use case hiding in everyday life. It points to the possibility of better mobile connectivity as an AI story that most people would only notice when everything is working well. The hosts frame the topic around AI-RAN, short for artificial intelligence in the radio access network. They describe the radio access network as the front line of mobile communications, where antennas and cell sites connect phones and other devices to the wider network. Andreas Oedemann joins from the Rode & Schwarz R&D team as an electrical engineer and AI-RAN expert. The conversation starts with a straightforward request for the general idea behind AI-RAN. From there, it moves into the mechanics of how mobile systems understand the channel between a phone and a base station. The guest explains that smartphones send pilot signals that are known to both ends of the link. Those pilots sit between the data being transmitted and are not the user’s actual message or call. Their role is to help the base station figure out how the signal was transmitted and how it was distorted while traveling through the air. The episode emphasizes that these signals are defined by specification, so every base station and smartphone understands the same structure. It notes that channel estimation today is still done with handcrafted algorithms. The hosts then raise the possibility that artificial intelligence could learn those patterns instead. The appeal is not a universal one-size-fits-all model. The appeal is a method that can adapt to the specific environment where infrastructure is deployed. The conversation uses that point to show why local conditions matter so much in mobile networking. Buildings, trees, and moving objects all affect the channel differently. The hosts compare that reality with the idea of tailoring algorithms to a mountain area versus a flatter one. The guest agrees that the best case would be highly tailored systems, while also noting that cost and benefit have to justify that level of engineering effort. The episode suggests that 5G today is still a very flat problem space. That makes the leap to more adaptive AI less automatic than it might first sound. It also keeps the discussion grounded in practical deployment rather than abstract promise. Near the end, the framing expands from AI for RAN to AI on RAN. That distinction covers both using AI to improve the infrastructure itself and supporting AI applications that run on top of it. The closing remarks return to the everyday value of that infrastructure. The show presents mobile connectivity as something people rely on constantly, even when they rarely think about it. It ends by inviting listeners to use the show notes if they want to go deeper. The overall result is a steady, clear explanation of how AI could reduce friction in the networks behind ordinary phone use.

As heard by us

A grounded look at AI-RAN and how it could improve mobile networks.

This episode leans into the practical side of AI, centering on AI-RAN and the modest but real promise of better phone connections, fewer dropped calls, and less buffering.

Read the full review in PlayNext →

Why you'd press play

Want AI explained as better phone service, not a buzzword?

Read the full recommendation in PlayNext →

Listen to the show on

Apple Podcasts Spotify