| Paper | Title | Page |
|---|---|---|
| WEP027 | System Integration Effort on MagViz a Liquid Explosive Detection Device | 462 |
|
||
|
Funding: Funded by the U. S. Department of Homeland Security. LA-UR- 09-02350 Los Alamos National Laboratory (LANL) technologists have adapted Magnetic Resonance Imaging (MRI) technology from medical devices to create MagViz, a new diagnostic and imaging device distinguishing potential-threat liquids from the harmless substances that the public might take onboard an aircraft. The MagViz system is a new ultra-low-field MRI approach using (SQUIDs) Superconducting Quantum Interference Devices for data acquisition. Linked with a computer database, MagViz reliably identifies some 50 liquids from the chemical fingerprints of their dynamic response to polarization stimulus wavefoms. The underlying technology for MagViz has been developed for an experimental laboratory environment. A controls team took the individually developed, highly incoherant, multi-platform collection of hardware and software applications, and integrated them into a coherant system. This paper will describe the integration effort, the problems faced, and the lessons learned when integrating a large number of hetterogenous systems; I.e. polarization stiumulus waveform generation system, SQUID control system, conveyer belt, camera image acquisition, data analysis library, and user interface. |
||
| THP083 | Lessons Learned Enhancing EPICS CA for LANSCE Timed and Flavored Data | 835 |
|
||
|
Funding: Work supported by US Department of Energy under contract DE-AC52-06NA25396. A previous paper described an upgrade to EPICS enabling client side tools at LANSCE to receive subscription updates filtered selectively to match a logical configuration of LANSCE beam gates, as configured by a user in the control room. To accommodate the upgrade fundamental changes were made in the EPICS core components. First, the event queue in the EPICS server was upgraded to buffer record (function block) and device specific parameters accessed generically via new software interfaces for introspection of 3rd party data. In contrast, the event queue in previous versions of EPICS was strictly limited to buffering only value, timestamp, and alarm status tuples. Second, the Channel Access server was upgraded to filter subscription updates. To make the filtering agent very flexible at runtime, filtering is accomplished by runtime interpretation of byte code generated from filtering expressions provided by the subscribing client. In this follow on paper, the performance of the new system, our future plans, and the lessons learned during this software development project will be described. |