SERPent:用于e-MERLIN的自动化减少和RFI缓解软件
SERPent: Automated reduction and RFI-mitigation software for e-MERLIN
摘要:
用于干涉测量的脚本化E-merlin RFI缓解PipelinE(SERPent)是一种自动化的减少和RFI缓解程序,利用了最初为LOFAR管道开发的SumThreshold方法(Offringa等人,2010 b)。SERPent是用Parseltongue语言编写的,能够与天文图像处理软件(AIPS)程序进行交互。此外,SERPent是一个简单的“开箱即用”Python脚本,它易于设置,并且没有编译器。除了标记受RFI影响的vibrations之外,脚本还标记天线零幅度下降和e-MERLIN系统固有的Lovell望远镜相位校准器固定扫描。这里介绍了SERPent的标记和计算性能,用于L波段(1.3 - 1.8 GHz)和C波段(4 - 8 GHz)观测的e-MERLIN调试数据集。对于更有问题的L波段观测,RFI通常为< 20 - 25%,而对于通常RFI更安静的C波段,RFI通常为< 5%。RFI检测和标记的水平比视觉手动标记更准确和微妙,输出立即准备用于AIPS校准。SERPent是完全并行的,并已在一系列计算系统上进行了测试。当前的标记率在“高端”计算机(16个CPU,100 GB内存)上为110 GB/天,相当于约6.9 GB/CPU/天,当e-MERLIN完成调试时,性能有望提高。自动化归约和校准程序的完善对于e-MERLIN遗留项目和未来的干涉仪(如SKA和相关的探路者(MeerKAT和ASKAP))至关重要,其中庞大的数据大小(> TB)使传统的天文学家交互变得不可行。
英文摘要:
The Scripted E-merlin Rfi-mitigation PipelinE for iNTerferometry (SERPent) is an automated reduction and RFI-mitigation procedure utilising the SumThreshold methodology (Offringa et al. 2010b), originally developed for the LOFAR pipeline. SERPent is written in the Parseltongue language enabling interaction with the Astronomical Image Processing Software (AIPS) program. Moreover, SERPent is a simple "out of the box" Python script, which is easy to set up and is free of compilers. In addition to the flagging of RFI affected visibilities, the script also flags antenna zero-amplitude dropouts and Lovell telescope phase calibrator stationary scans inherent to the e-MERLIN system. Both the flagging and computational performances of SERPent are presented here, for e-MERLIN commissioning datasets for both L-band (1.3 - 1.8 GHz) and C-band (4 - 8 GHz) observations. RFI typically amounts to < 20 - 25% for the more problematic L-band observations and < 5% for the generally RFI quieter C-band. The level of RFI detection and flagging is more accurate and delicate than visual manual flagging, with the output immediately ready for AIPS calibration. SERPent is fully parallelised and has been tested on a range of computing systems. The current flagging rate is at 110 GB/day on a "high-end" computer (16 CPUs, 100 GB memory) which amounts to ~ 6.9 GB/CPU/day, with an expected increase in performance when e-MERLIN has completed its commissioning. The refining of automated reduction and calibration procedures is essential for the e-MERLIN legacy projects and future interferometers such as the SKA and the associated pathfinders (MeerKAT and ASKAP), where the vast data sizes (> TB) make traditional astronomer interactions unfeasible.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。