- South Korean researchers have created Concretess, a new way of improving wireless communication
- Concretess avoids large cod books, improves image transmission and reduces errors
- The team believes that it could help 6G networks, smart factories and health devices
Researchers in South Korea have developed a new approach to semantic communication that could make future wireless systems faster and more efficient.
Concretesc’s new method was created by a team led by Dr Dong Jin Ji, Associate Professor at the National University of Science and Technology in Seoul, and was published on June 19, 2025 IEEE wireless communication letters.
Semantic communication is a change in wireless technology where meaning is sent rather than raw data. For example, when transmitting an image, the system prioritizes what the image represents instead of sending exactly each pixel. This saves time and bandwidth and could be particularly useful for artificial intelligence and connected devices.
Potential for 6G
Existing systems often depend on vector quantification, a process that uses giant “cod books” to store possible signal models. These codes are not only heavy to manage but fight with errors and noise.
Concrete resolves this with a different mathematical idea.
Despite its name, it has nothing to do with the material used in buildings. Instead, “concrete” here refers to a special probability distribution in automatic learning.
This tool makes it possible to transform continuous information into digital signals more easily, allowing the system to generate the bits which it needs directly, without the management of large codes of codes.
“Unlike the quantification of vectors (VQ) – a cutting -edge digitization technique that suffers from the noise of the channels and the divergence of codes of codes during training – our frame offers a fully differentiaiable solution to quantification, allowing end -to -end training even under the noise of the canal,” said Dr. Ji.
“In particular, due to the nature of the Cantes which directly generate the required bits flow, it is possible to form a pair of multiple length models in a relatively simple masking scheme,” he added.
In the simulations, the researchers claim that Cannetestc has surpassed the methods based on VQ in the structural similarity and the peak signal / noise ratio. It has also reduced complexity, because its operations only increase with a length of bit rather than developing exponentially with the size of the codes.
Researchers think that this framework could play an interesting role in new generation wireless systems, such as 6G.
They cite other potential uses such as intelligent factories with ultra-dense machine communications, as well as health care and lifestyle monitoring systems powered by small AI compatible devices.
