作者: Djellel Eddine Difallah , Philippe Cudre-Mauroux , Sean A. McKenna
DOI: 10.1109/MIC.2013.84
关键词: Network topology 、 Anomaly detection 、 Real-time computing 、 Computer network 、 Array processing 、 Computer science 、 Smart city 、 Context (language use) 、 Scalability 、 Wireless sensor network 、 Network architecture
摘要: Dynamically detecting anomalies can be difficult in very large-scale infrastructure networks. The authors' approach addresses spatiotemporal anomaly detection a smarter city context with large numbers of sensors deployed. They propose scalable, hybrid Internet for dynamically potential real time using stream processing. enables analytically inspecting and comparing globally array Deployed on pipe network topology 1,891 nodes, this effectively detect characterize while minimizing the amount data shared across network.