Collaborative Research: Large-Aperture Experiment to Detect the Dark Age (LEDA)

技术摘要探测黑暗时代的大口径实验(LEDA)项目旨在探测大爆炸(红移16-40)约1亿年后星系间介质中中性氢(21厘米静止波长)的超精细发射。探测将对结构形成模型和宇宙中第一批恒星和黑洞的形成提供第一个观测约束。LEDA将开发信号处理仪器，并将其集成到长波长阵列(LWA)的新第一站中。这包括一个大N相关器，服务于LWA-1的所有512个偶极天线，利用分组化的Casper架构，并将F级和X级的FPGA和GPU结合在一起。迭代校准和成像将依赖于扭曲快照成像，并取自支持GPU的库(CUWARP)，该库专为支持使用固定偶极阵列的广域全极化成像而设计。校准技术将包括对电离层折射和方向相关偶极子增益的校正，以及探索脉冲星数据分析以提高性能。准确的校准和成像将是LEDA的关键要求，这是减去明亮的前景天空和探测微弱的中性氢信号所必需的。从计算的角度来看，LEDA是O(100)teraflop/s的挑战，它支持一种可扩展的架构，着眼于开发需要高能效的10petaflop/s性能的无线电阵列。再电离阵列氢纪元(HERA2)的第二阶段就是一个例子。这些恒星预计将比今天我们周围的恒星质量大得多。超大质量黑洞是在同一时间、更早还是更晚形成的？今天宇宙学最大的挑战之一就是研究这些第一代天体。人们普遍假设，它们的形成始于大爆炸后约1亿年，但目前还没有数据来检验这一理论。在如此年轻的年龄(仅为今天的1%)研究宇宙的唯一可用的手段是通过恒星和黑洞之间的星系间介质的电磁辐射。今天，这是热的和电离的等离子体，但在早期宇宙中，它是一个巨大的冷中性氢气储存库，滋养了第一批恒星和黑洞的形成，并大量辐射长波辐射。LEDA项目试图应用前沿射电天文技术来首次探测到这种信号。LEDA将建造一台“无线电相机”，用于部署在新墨西哥州的射电望远镜长波阵列上，该望远镜的第一个100米直径的口径最近刚刚完成。LEDA相机将结合几项创新技术和数据分析技术，让学生和年轻科学家有机会加入尖端科学和开发最先进的工具。特别是，LEDA将利用图形处理单元(GPU)的巨大计算能力和灵活性-为视频游戏提供动力-在长达10米的波长(比可见光辐射长1000万倍)的几乎整个天空的瞬时图像。从这些图像中，我们和其他星系的光将被高精度地减去，从而能够搜索宇宙黎明时的信号。LEDA将推动宇宙学的前沿，同时为未来的射电天文设施做出贡献，在那里，大规模的计算和信号处理系统将是关键。LEDA工作的跨学科成果将使天文、计算和太阳科学受益。

Technical AbstractThe Large Aperture Experiment to Detect the Dark Ages (LEDA) project seeks to detect hyperfine emission from neutral Hydrogen (21 cm rest wavelength) in the intergalactic medium about 100 million years after the Big Bang (redshifts 16-40). A detection would deliver the first observational constraints on models of structure formation and on the formation of the first stars and black holes in the Universe. LEDA will develop and integrate signal processing instrumentation into the new first station of the Long Wavelength Array (LWA). This comprises a large-N correlator serving all 512 dipole antennas of the LWA-1, leveraging a packetized CASPER architecture and combining FPGAs and GPUs for the F and X stages. Iterative calibration and imaging will rely on warped snapshot imaging and be drawn from a GPU-enabled library (CUWARP) that is designed specifically to support wide-field full polarization imaging with fixed dipole arrays. Calibration techniques will include correction for ionospheric refraction and direction dependent dipole gains, and exploration of pulsar data analysis to improve performance. Accurate calibration and imaging will be crucial requirements for LEDA, necessary to subtract the bright foreground sky and detect the faint neutral Hydrogen signal. From the computational standpoint, LEDA is a O(100) TeraFlop per second challenge that enables a scalable architecture looking toward development of radio arrays requiring power efficient 10 PetaFlop per second performance. Stage two of the Hydrogen Epoch of Reionization Array (HERA2) is one example.Lay AbstractWhen did the first stars form? These stars are expected to be much more massive than the stars that are around us today. Did supermassive black holes form at the same time, earlier, or later? One of the great challenges of cosmology today is the study of these first generation objects. Their formation is widely hypothesized to have begun about 100 million years after the Big Bang, but no data are available to test this theory. The only available means to study the Universe at so young an age (just 1% of what it is today) is via electromagnetic radiation from the intergalactic medium between the stars and black holes. Today, this is hot and ionized plasma, but in the early Universe it was a vast reservoir of cold neutral Hydrogen gas that fed the formation of the first stars and black holes and radiated long wavelength radiation copiously.The LEDA project seeks to apply frontier radio astronomical techniques to make the first detection of this signal. LEDA will build a "radio camera" for deployment to the Long Wavelength Array, a radio telescope in New Mexico whose first 100m-diameter aperture was recently completed. The LEDA camera will combine several innovative technologies and data analysis techniques, giving students and young scientists the opportunity to join in cutting-edge science and development of the most advanced tools. In particular, LEDA will harness the massive computing power and flexibility of Graphics Processing Units (GPUs) - the engines that power video games - to make instantaneous images of nearly the whole sky at up to 10 meters wavelength (10 million times longer than visible radiation).From these images the light of our and other galaxies will be subtracted with high accuracy, enabling a search for signals from the dawn of the Universe. LEDA will push the frontiers of cosmology while contributing groundwork for future radio astronomical facilities where massive computing and signal processing systems will be lynchpins. Cross-disciplinary outgrowths of the LEDA effort will benefit astronomical, computational and solar sciences.