AI. Kind of confusing isn’t it? Even as we all find it embedded into virtually every digital and electronic process and technology that we encounter, we also learn that it is going to be the cause of a large number of problems. Serious ones.
In no particular order, mainstream media has recently warned that the uptake of AI, especially its in agentic form, will be the driver of a massive construction programme for massive data centres that will be a massive drain on resources such as electricity and water; the cause of a huge economic crisis, worse than the banking crash of 2008, as the AI industry’s financial bubble bursts; be the source of bio attacks; and more or less cause the end of life as we know it and the destruction of humanity.
Apart from all of that, the picture looks bright. Yet actually, in the world of broadcasting, the picture could be literally very bright. Indeed, research from Waseda University in Japan is suggesting that AI can be used as an essential tool in ensuring quality control for 5G multicast broadcasting.
Waseda University is an institution in the Japanese education system. Formed in 1882 and located in the heart of Tokyo, the private research university has had a long track record in being dedicated to academic research and civic engagement at both the local and global levels.
With alumni including nine prime ministers and leaders in business, science and technology, Waseda has strong collaborations with overseas research institutions and has the stated aim of being committed to advancing what it calls cutting-edge research. Helping to achieve reliable last-mile delivery of cable television through the use of AI-enhanced 5G wireless communications is among the latest projects looking to maintain this tradition.
And this project does not just have technological aims and ambitions; it also sets out to bridge what could be a digital accessibility gap. That is, delivering HD television over local 5G networks to help modernise cable TV services in buildings for which installing fixed assets – such as fibre-optic cables – is impractical. And in doing so, using AI to ensure more reliable video delivery in more places.
One of the key founding elements of the research – led by Professor Jiro Katto and PhD candidate Kasidis Arunruangsirilert – was how cable television was a cornerstone of how national broadcasts and emergency information reach thousands of households in Japan.
Yet looking as to how all buildings and homes could receive such transmissions, Katto and Arunruangsirilert suggested that as many older buildings were fundamentally not built with fibre optics in mind, retrofitting them with the necessary wiring for fixed cable TV would be too costly or impractical. This they said would leave residents unable to access wired broadcast services or depend on wireless services.
However, despite Japan boasting one of the most advanced and widely distributed 5G wireless infrastructures in the world, the research highlighted how multicast broadcasting over 5G has a fundamental limitation. That is, ordinary 5G connections on smartphones are two-way, so when a data packet fails to arrive, the receiving device can simply request retransmission. Multicast broadcasting may not support such return channels, which means lost data packets cannot be resent, causing the video stream to freeze momentarily.
Ultimately, the research highlighted how conventional 5G communication protocols, designed around retransmission requests and maximising speed, are not built to handle situations where retransmission is limited. And since lost data cannot be retransmitted in 5G multicast broadcasting, temporary video freezing could be a common experience for viewers.
The proposed solution to address these issues was a lightweight AI model designed to predict wireless conditions before they deteriorate and to adjust transmission settings proactively to enable stable, real-time video delivery via 5G broadcasting. At the heart of this is 5G MBS (Multicast/Broadcast Services), a technology that uses 5G wireless signals to deliver the same television stream to many households simultaneously.
By making efficient use of the limited radio spectrum, 5G MBS offers a realistic path to modernising the fixed infrastructure. Moreover, its use could help bridge the information gap in underserved areas, contributing to a society where everyone has equal access to broadcast services.
Highlighting the crucial nature of AI in the solution, the research team revealed that they used an AI model trained using approximately 26 million measurements collected every 0.5 milliseconds from a commercial 5G network. This was intended to capture rapid fluctuations in radio conditions that standard “coarse” datasets could miss. And in what was regarded as a crucial factor, the model relied entirely on information already collected by smartphones during standard operation. This was considered compact enough to run in real time on consumer devices without requiring specialised hardware.
In specific detail, the tests on a real-world commercial 5G network showed that the AI model selected an error-free transmission setting for approximately 87 per cent of video segments, compared with only 32 per cent for a conventional speed-oriented approach. The model also operated in less than 0.07 milliseconds on smartphone chipsets released from 2020 onward, introducing no perceptible delay to viewers.
Commenting on the results, Katto and Arunruangsirilert said the study stood as a practical example of AI-native wireless communication in which AI takes on decision-making responsibilities in next-generation networks. They added that the research now represents a meaningful step toward the long-sought convergence of broadcasting and broadband on a single, spectrally efficient wireless platform.
Looking outside the box, to coin a TV colloquialism, the researchers see their work as demonstrating not only how not how wireless television services could be improved and made more generally available, but also how local 5G infrastructure, currently used primarily by large organisations, could deliver practical public services. “We hope that this work encourages broader use of local 5G as a tool for solving social challenges and serving the public good,” they said.
In addition, they suggested that their work could also potentially benefit non-TV one-way wireless communication use cases where retransmission is impossible. This could include satellite communications, autonomous vehicles, industrial systems and general scientific exploration.
So, in conclusion, there are places where AI could be hugely beneficial and not a threat to life on Earth. In Japan it could ensure traditionally underserved homes can be reliably informed and entertained. AI is truly revolutionary and this particular revolution will definitely be televised.