论文标题
下一代地面检测器的模拟数据研究：由于混乱噪声而引起的匹配过滤的性能损失
Mock data study for next-generation ground-based detectors: The performance loss of matched filtering due to correlated confusion noise
论文作者
论文摘要
下一代（3G/XG）地面重力波（GW）探测器，例如爱因斯坦望远镜（ET）和宇宙资源管理器（CE）将在未来十年开始观察。由于这些探测器的灵敏度极高，将观察到整个宇宙中的大多数恒星质量紧凑型二元合并。还可以预期，3G探测器的敏感性将降至2-7 Hz。观察到的二进制中子星信号的持续时间可能会增加到几个小时或几天。信号的丰度和持续时间会导致它们随着时间的流逝而重叠，这可能会形成混乱的噪声，在使用幼稚匹配的过滤时可能会影响单个GW源的检测；匹配的过滤仅是固定高斯噪声的最佳选择。我们使用GWTC-3目录告知的最新人群模型为CE和ET创建模拟数据，并研究由于重叠信号而导致的匹配过滤的性能损失。我们发现性能损失主要来自噪声测得的振幅光谱密度的偏差。 CE（ET）的红移覆盖范围可以减少15％-38％（8％-21％），具体取决于合并率估计值。与仪器噪声的贡献相比，混淆噪声对总信噪比（SNR）的直接贡献通常可以忽略不计。我们还发现，相关的混淆噪声对ET网络SNR的正交总和规则具有可忽略的影响，但如果未应用缓解，包括CE在内的检测器网络的高探测器框架质量信号的网络SNR（包括CE）。对于ET，无效流可以减轻天体物理的前景。对于CE，我们证明了一种计算高效，直接的单探测器信号减法方法将总噪声抑制至几乎仪器噪声水平。这将允许近距离搜索。
The next-generation (3G/XG) ground-based gravitational-wave (GW) detectors such as Einstein Telescope (ET) and Cosmic Explorer (CE) will begin observing in the next decade. Due to the extremely high sensitivity of these detectors, the majority of stellar-mass compact-binary mergers in the entire Universe will be observed. It is also expected that 3G detectors will have significant sensitivity down to 2-7 Hz; the observed duration of binary neutron star signals could increase to several hours or days. The abundance and duration of signals will cause them to overlap in time, which may form a confusion noise that could affect the detection of individual GW sources when using naive matched filtering; matched filtering is only optimal for stationary Gaussian noise. We create mock data for CE and ET using the latest population models informed by the GWTC-3 catalog and investigate the performance loss of matched filtering due to overlapping signals. We find the performance loss mainly comes from a deviation in the noise's measured amplitude spectral density. The redshift reach of CE (ET) can be reduced by 15%-38% (8%-21%) depending on the merger rate estimate. The direct contribution of confusion noise to the total signal-to-noise ratio (SNR) is generally negligible compared to the contribution from instrumental noise. We also find that correlated confusion noise has a negligible effect on the quadrature summation rule of network SNR for ET, but might reduce the network SNR of high detector-frame mass signals for detector networks including CE if no mitigation is applied. For ET, the null stream can mitigate the astrophysical foreground. For CE, we demonstrate that a computationally efficient, straightforward single-detector signal subtraction method suppresses the total noise to almost the instrument noise level; this will allow for near-optimal searches.
