Ultra-Low-Latency Edge Inference for Distributed Sensing with Short Packets

Artificial intelligence (AI) is expected to be a defining component in sixth-generation (6G) wireless networks. One specific use is AI-empowered sensing, where sensor data will be processed at the network edge, giving rise to integrated sensing and edge AI (ISEA). Many sensing applications, such as autonomous driving and industrial manufacturing, are latencysensitive and require end-to-end (E2E) performance guarantees under stringent deadlines. However, data-agnostic ultrareliable and low-latency communication (URLLC) techniques designed for 5G fall short in achieving the optimal E2E sensing performance. In this work, we introduce an ultra-low-latency (ultra-LoLa) inference framework for perceptive networks that facilitates the analysis of E2E sensing accuracy in distributed sensing by jointly considering communication reliability and inference accuracy. By characterizing the tradeoff between packet length and the number of sensing observations, we derive an efficient optimization procedure that closely approximates the optimal balance. The experimental results show that the proposed approach outperforms conventional reliability-oriented protocols with respect to sensing performance under a latency constraint.