Smart cities, Industry 4.0, tele-surgery, e-health, volumetric video streaming, artificial intelligence. With the rapid development of digitalisation across various sectors, a wide range of applications have emerged that demand unprecedented levels of reliable, low latency connectivity and timely decision-making.

Just looking at data traffic alone, the amount of available and generated data increases exponentially each year. Every day, approx. 400 billion gigabytes of data is generated worldwide.

Communicating, processing, and storing such vast amount of data is highly resource-demanding, and current solutions and technologies are not sustainable. As a result, researchers and engineers are striving to develop more energy-efficient, scalable, and sustainable data communication and computing technologies.

One of these researchers is Qi Zhang, a newly appointed full professor at Aarhus University:

"Enormous amount of data are being generated every day. But the problem isn’t just about the sheer volume - it’s also about the diversity of data types. While just a few years ago, data mainly consisted of text, audio, video, and IoT time series, now we want to transmit 3D data, control signals, haptic feedback, and even data of AI models, each with its own characteristics and requirements. Another important question is how to efficiently leverage data for various applications, such as extracting knowledge and learning from the data. These present significant challenges for communication systems, networks and data computing infrastructure," she says.

Qi Zhang researches Edge Intelligence, Internet of Things, data compression and analytics at the Department of Electrical and Computer Engineering and is co-leading the research group NeXt (Network Computing, Communications and Storage).

She believes that integrating Edge Intelligence into data communication and computing systems can provide holistic and efficient solutions to the challenges.

"Today’s communication systems are built based on Shannon’s Theory proposed in the 1940s, which focuses on reliable transmission of bit sequences, and does not consider the meaning of the transmitted data. It is seen as a ‘transmit then understand’ approach. An example of this could be, that after a receiver has received data of an image, the image can then be understood. This is the traditional way of data communication. It is anticipated that the existing communication systems will soon reach a resource bottleneck, due to its limitations in addressing the new challenges " she says.

In many emerging applications, the ultimate data consumers are AI-enabled machines, such as those performing tasks like image classification. Many machine intelligence services are task-oriented, meaning they do not require all bits to be transmitted reliably, but rather only the information relevant to completing the specific task.

Qi Zhang continues:

“Furthermore, with the advancement of Edge Intelligence, the transmitter will be able to understand the meaning of data and only transmit the necessary data, which is seen as an ‘understand before transmission’ process. This is a novel paradigm shift in communication, known as Goal-oriented Semantic Communication, which not only improves communication efficiency but also enhances computing for downstream tasks. For example, our team recently developed a semantic-aware data compression that efficiently extracts semantic meaning inherent in data series. By preserving the essential patterns, it enables direct data analytics, such as outlier detection, without data reconstruction. Interestingly, experimental results demonstrated that analytics on the compacted data yields even higher accuracy at reduced response time. This holds great potential for driving digitalisation in a more cost-effective and sustainable way.”

Research career in Denmark

Qi Zhang was born and raised in Ningbo, China. She completed her Bachelor's degree in China before taking her Master's degree and PhD at DTU in Denmark. After completing her PhD, she worked for two years as a consulting engineer in Copenhagen before joining the Engineering College of Aarhus which later merged with Aarhus University.

"I didn't always dream of becoming a researcher, but whenever I see a challenging problem, I want to solve it. My Master’s thesis supervisor encouraged me to apply for a DTU PhD scholarship, and that's how my research career started. I started off as a young Master’s student in Denmark, and now I'm a professor," she says.

Qi Zhang has led several national and international research projects at the Department of Electrical and Computer Engineering, for example, the TOAST project, which aims to develop technology and scale up talent for developing the Tactile Internet (the Internet of Skills) - a key component of the future digital world. The TOAST project has received EUR 2.7 million (DKK 20.4 million) in funding from Horizon Europe's Marie Skłodowska-Curie Actions programme.

Other examples are: the AgilE-IoT project, which develops agile edge intelligence for time-critical IoT applications; Light-IoT which designs novel data compression and IoT framework that allows data analytics directly on compressed data; and eTouch, which develops Edge Intelligence for immersive telerobotics for communication over large distance, are funded by Independent Research Fund Denmark; NUEI (Nordic University Collaboration on Edge Intelligence), which is a strategic research collaboration project of four Nordic Universities funded by NordForsk.

Professor Qi Zhang will give her inaugural lecture on 12 August 2025.

Contact

Professor Qi Zhang
Aarhus University, Department of Electrical and Computer Engineering
​​​​​​Mail: qz@ece.au.dk
Tel.: +4541893253