Collaborative Edge AI Inference over Cloud-RAN

arXiv:2404.06007v1 Announce Type: cross
Abstract: In this paper, a cloud radio access network (Cloud-RAN) based collaborative edge AI inference architecture is proposed. Specifically, geographically distributed devices capture real-time noise-corrupted sensory data samples and extract the noisy local feature vectors, which are then aggregated at each remote radio head (RRH) to suppress sensing noise. To realize efficient uplink feature aggregation, we allow each RRH receives local feature vectors from all devices over the same resource blocks simultaneously by leveraging an over-the-air …

abstract architecture arxiv cloud collaborative cs.ai cs.it cs.lg data devices distributed edge edge ai eess.sp extract feature head inference math.it network noise paper radio ran real-time samples sensing sensory type vectors