III: Small: High Throughput Real-Time Astronomy: Discrimiation, Dissemination and Decision

点击翻译按钮获取中文摘要

This proposal will be awarded using funds made available by the American Recovery and Reinvestment Act of 2009 (Public Law 111-5).Intellectual Merit:We propose to build an event-driven network, a continuation of the NSF VOEventNet, applying the event-driven technology to Astronomy. There are many large surveys coming online (CRTS, PTF, LOFAR, LIGO, LSST, SKA etc) that are scanning the sky repeatedly, to find what is changing and why. Observation can be caused by supernova, gamma-ray burst, blazar eruption, planetary microlensing, or other exciting astrophysics. In many cases, rapid dissemination and follow-up observation is the key to discovery, to get magnitudes and multi-wavelength coverage.The new VOEventNet will ingest events from multiple event surveys, professional and amateur and disseminate them for free at all levels, from child to major observatory. Trigger (selection) criteria can be built precisely and individually. When an event arrives, these triggers (boolean functions) are run to decide if an action is to happen or not. These triggers will work with heterogeneous, multi-sourced data. Actions can include the moving and operating of a sensor, either human or robot guided, sending messages, fetching data to build a data portfolio, and running classification rules. Further, actions can cause the generation of new events, thus inducing a workflow. Events come in real time from an automated pipeline of a major sky survey or from a small college observatory or amateur astronomer doing follow-up observations. These authors have previously registered the semantic meaning of the parameters that will be used in the actual events, so that automated systems can be effective, real-time, semantically aware decision-makers. The new VOEventNet will use an international standard (VOEvent) to be part of the global event infrastructure, exchanging events with other event brokers, such as NASA's GCN.Broader Impact:We propose here a cyber-infrastructure for events that are created from real-time sensors. The sensor could be many things, for example a radiation detector at a port. Excess click rate of the Geiger counter creates an event that is automatically assembled with other data, such as country of origin, and decisions made, perhaps using an additional sensor, the results triggering another action. Only important events merit human attention. This "Sensor-Event-Action" pattern occurs in many other fields of science and engineering, for example early earthquake warning, network provisioning when an internet story becomes viral, leak or pollution detection, and many others.