AI Optimizes Antenna Performance in Real-World Conditions

Anritsu Corporation has jointly verified AI-based antenna optimization technologies with SK Telecom, POSTECH, and Bluetest, demonstrating performance improvements under real user conditions. The results, presented at MWC Barcelona 2026, highlight a data-driven approach to optimizing antenna systems in 5G networks.

Measurement-Based Optimization in Real Environments
The verification was conducted using over-the-air (OTA) measurements to capture realistic user scenarios, including free-space operation, handheld usage, and head-proximate conditions. These scenarios introduce variability in radio frequency (RF) performance due to user interaction and environmental factors.

Measurement data included throughput and error correlation coefficient (ECC), enabling quantitative evaluation of antenna behavior under dynamic conditions.

AI-Driven Antenna Configuration
Using AI-based analysis, the system modeled RF performance variations across different antenna tuner states. Based on these models, optimal antenna switching configurations were identified automatically.

This approach enables dynamic antenna optimization, maintaining communication quality by adapting to changing RF conditions in real time.

Performance Improvements in MIMO Systems
The verification demonstrated measurable throughput gains across multiple configurations. In an 8Rx (eight-receive-antenna) setup, significant improvements were observed across user scenarios. In a 4Tx (four-transmit-antenna) configuration, throughput increased by more than two times in certain conditions.

These results indicate that AI-driven optimization can enhance spectral efficiency and link performance in complex multi-antenna systems.

Test Platforms and Data Acquisition
The verification utilized Anritsu’s MT8000A 5G NR test platform and MT8870A wireless measurement system.

The MT8000A enabled MIMO signal generation and throughput evaluation under controlled 5G NR conditions, supporting configurations such as 4×4 and 8×8 MIMO. It provided synchronized multi-port measurements and high-precision data acquisition.

The MT8870A platform was used for RF characteristic analysis across antenna paths, measuring power and performance variations under different tuner states. These measurements formed the input data for AI-based modeling and optimization.

Workflow for Data-Driven Antenna Optimization
The verification established a workflow that integrates measurement data with AI analysis. This includes evaluating RF performance across user scenarios, comparing antenna path characteristics, deriving optimal switching states, and validating performance improvements.

This method moves beyond static antenna design by incorporating real-world conditions into the optimization process.

Relevance for 5G and Wireless Systems
As 5G networks rely on advanced MIMO configurations, antenna performance becomes increasingly sensitive to user interaction and environmental variability. AI-driven optimization provides a mechanism to adapt antenna behavior dynamically, improving throughput and reliability.

Within a digital supply chain for telecommunications, such technologies support more efficient network performance, better user experience, and optimized use of spectrum resources.

Edited by Romila DSilva, Induportals Editor, with AI assistance.