Agentic semantic sensing cuts costs adapting radio network sensing
Agentic Semantic Sensing for Resource-Adaptive AI-RAN
Artificial Intelligence
Summary
Collecting only the information needed for a task instead of gathering everything can save time and resources. The authors propose a system that decides what to sense and when during operation, based on what has already been observed. This adaptive approach uses a type of AI model to update knowledge and decide the best sensing strategy, balancing value and cost. Tests showed it reduces sensing cost significantly while keeping accuracy high.
What this means in practice
- •For network schedulers: Optimize radio sensing schedules dynamically to reduce cost while maintaining communication task performance in AI-driven radio networks.
- •For wireless system engineers: Implement adaptive sensing strategies in wireless systems that respond to task progress and save sensing resources.
Authors
Zhongqin Wang, Xiaoqi Zhang, Nan Yang, Kai Wu, J. Andrew Zhang, Y. Jay Guo
Abstract
Semantic sensing (SemS) acquires task-relevant information rather than reconstructing complete physical information. Existing SemS formulations typically operate open loop: sensing configurations and observation schedules are fixed before inference and cannot respond to evolving task-level evidence. We propose Agentic SemS, a closed-loop framework for AI-enabled radio access networks (AI-RANs) that controls sensing within a communication-feasible profile set. A profile-conditioned causal Transformer updates the semantic belief from streaming observations, while key-value caching enables efficient state updates across profile changes without repeatedly processing the complete history. A semantic utility network estimates the task-level benefit of acquiring the next observation block under each feasible profile after accounting for sensing cost. The resulting continuation utilities jointly support next-profile selection and semantic early exit, adapting sensing configuration and duration to evolving evidence. The expected semantic gain is further related to conditional mutual information, providing a value-of-information interpretation of continued online sensing. Experiments on Widar3.0 with six emulated sensing profiles show that, in comparison with full-sequence High, the resource-efficient Agentic setting reduces normalized cumulative sensing cost by 25.33% while achieving 85.79% Macro-F1. At the same utility checkpoint, semantic early exit provides a further 12.35% cost reduction over adaptive sensing without early exit, with a 0.97-percentage-point Macro-F1 decrease.