Programmable and GPU-Accelerated Edge Inference for Real-Time ISAC on NVIDIA ARC-OTA

Abstract: The transition of cellular networks to (i) software-based systems on commodity hardware and (ii) platforms for services beyond connectivity introduces critical system-level challenges. As sensing emerges as a key feature toward 6G standardization, supporting Integrated Sensing and Communication (ISAC) with limited bandwidth and piggybacking on communication signals, while maintaining high reliability and performance, remains a fundamental challenge. In this paper, we provide two key contributions. First, we present a programmable, plug-and-play framework for processing PHY/MAC signals through real-time, GPU-accelerated AI applications on the edge Radio Access Network (RAN) infrastructure. Building on the Open RAN dApp architecture, the framework interfaces with a GPU-accelerated gNB based on NVIDIA ARC-OTA, feeding PHY/MAC data to custom AI logic with latency under 0.5 ms for complex channel state information extraction. Second, we demonstrate the framework's capabilities through cuSense, an indoor localization dApp that consumes uplink DMRS channel estimates, removes static multipath components, and runs a neural network to infer the position of a moving person. Evaluated on a 3GPP-compliant 5G NR deployment, cuSense achieves a mean localization error of 77 cm, with 75% of predictions falling within 1 meter. This is without dedicated sensing hardware or modifications to the RAN stack or signals. We plan to release both the framework and cuSense pipelines as open source, providing a reference design for future AI-native RANs and ISAC applications.