Exhibition and Contributions of ComNets Dortmund at IEEE ICC 2026

Intelligent Reflecting Surfaces (IRSs) are considered a promising technology for future 6G mmWave networks, enabling wireless signals to be steered and enhanced through programmable reflections. However, existing analytical models typically assume ideal far-field conditions and become increasingly inaccurate when transmitters and receivers operate at finite distances from large IRS deployments. To address this challenge, an established IRS reflection model was extended to account for the finite distances between transmitter, IRS, and receiver. By considering the individual reflection angles that arise under these conditions, the proposed approach provides a significantly more accurate description of large passive IRSs, particularly in radiative near-field scenarios. The model was validated through both electromagnetic simulations and laboratory measurements. Results show that it can predict reflection patterns at distances down to 50 cm with up to 46.5 percentage points higher correlation than conventional far-field models. In addition, the improved modeling accuracy enables optimized reflector designs that avoid near-field-induced signal nulls, increasing the minimum in-beam radar cross section by more than 30 dB in the evaluated scenario. These findings highlight the importance of accounting for finite-distance and near-field effects in future IRS-assisted 6G mmWave networks. Accurate propagation models will be essential for reliable connectivity predictions and the design of efficient IRS deployments across a wide range of practical operating conditions.

Future 6G networks are expected to enable immersive teleoperation applications that require both low latency and high reliability. This work investigates how different visual feedback mechanisms affect teleoperation performance under challenging network conditions. To this end, a demonstrator consisting of two industrial robot arms, a virtual air hockey environment, and a VR-based user interface was developed. Users controlled the robots via hand gestures while receiving feedback either through conventional video streaming or a semantic digital twin. A user study with 25 participants compared both approaches under ideal network conditions as well as under emulated packet loss and network outages. The results show that digital-twin-based teleoperation is significantly more resilient to network impairments than traditional video streaming. While degraded video transmissions suffered from severe artifacts and interruptions, the digital twin maintained stable and responsive operation. Participants achieved up to 70% higher offensive and defensive performance with the digital twin under impaired network conditions. At the same time, transmitting semantic state information reduced the required bandwidth from approximately 12 Mbit/s to 900 kbit/s. These findings highlight the potential of semantic communication and digital twins for resilient and resource-efficient teleoperation in future 6G networks. 

Presented Contributions

• K. Šabanović, H. Schippers, K. Heimann, C. Wietfeld, “QoE Benchmarking of Resilient 6G Teleoperation: VR Digital Twin vs. Video Streaming,” in IEEE International Conference on Communications (ICC), Glasgow, Scotland, May 2026. [pdf] [Details]
• S. Häger, M. Kaudewitz, C. Wietfeld, “Near-field Extension of Analytical Reflection Model for Large-scale mmWave IRS Geometries in 6G,” in IEEE International Conference on Communications (ICC), Glasgow, Scotland, May 2026. [pdf] [Details]
• P. Gautam, C. Arendt, S. Fricke, C. Bockelmann, A. Dekorsy, C. Wietfeld, “RASP: Reliability-Aware SINR Prediction for Realistic Industrial Subnetworks,” in IEEE International Conference on Communications (ICC), Glasgow, Scotland, May 2026. [pdf] [Details]

Acknowledgments
The presented works were supported by the German Federal Ministry of Research, Technology and Space (BMFTR) through the projects 6GEM (16KISK038), GEM-X (16KISS005), PANGOLIN Networks (16KIS2357), and the 6GEM+ transfer hub (16KIS2412).