L-DIT: A dApp for Live Detectability, Identifiability and Trackability for ASOs on the Behavioral Dynamics Blockchain
Authors:
Anirban Chowdhury,
Yasir Latif,
Moriba K. Jah,
Samya Bagchi
Abstract:
As the number of Anthropogenic Space Objects (ASOs) grows, there is an urgent need to ensure space safety, security, and sustainability (S3) for long-term space use. Currently, no globally effective method can quantify the safety, security, and Sustainability of all ASOs in orbit. Existing methods such as the Space Sustainability Rating (SSR) rely on volunteering private information to provide sus…
▽ More
As the number of Anthropogenic Space Objects (ASOs) grows, there is an urgent need to ensure space safety, security, and sustainability (S3) for long-term space use. Currently, no globally effective method can quantify the safety, security, and Sustainability of all ASOs in orbit. Existing methods such as the Space Sustainability Rating (SSR) rely on volunteering private information to provide sustainability ratings. However, the need for such sensitive data might prove to be a barrier to adoption for space entities. For effective comparison of ASOs, the rating mechanism should apply to all ASOs, even retroactively, so that the sustainability of a single ASO can be assessed holistically. Lastly, geopolitical boundaries and alignments play a crucial and limiting role in a volunteered rating system, limiting the space safety, security, and sustainability. This work presents a Live Detectability, Identifiability, and Trackability (L-DIT) score through a distributed app (dApp) built on top of the Behavioral Dynamics blockchain (BDB). The BDB chain is a space situational awareness (SSA) chain that provides verified and cross-checked ASO data from multiple sources. This unique combination of consensus-based information from BDB and permissionless access to data allows the DIT scoring method presented here to be applied to all ASOs. While the underlying BDB chain collects, filters, and validates SSA data from various open (and closed if available) sources, the L-DIT dApp consumes the data from the chain to provide L-DIT score that can contribute towards an operator's, manufacturer's, or owner's sustainability practices. Our dApp provides data for all ASOs, allowing their sustainability score to be compared against other ASOs, regardless of geopolitical alignments, providing business value to entities such as space insurance providers and enabling compliance validation and enforcement.
△ Less
Submitted 20 June, 2024; v1 submitted 28 April, 2024;
originally announced April 2024.
Simultaneous Multivariate Forecast of Space Weather Indices using Deep Neural Network Ensembles
Authors:
Bernard Benson,
Edward Brown,
Stefano Bonasera,
Giacomo Acciarini,
Jorge A. Pérez-Hernández,
Eric Sutton,
Moriba K. Jah,
Christopher Bridges,
Meng **,
Atılım Güneş Baydin
Abstract:
Solar radio flux along with geomagnetic indices are important indicators of solar activity and its effects. Extreme solar events such as flares and geomagnetic storms can negatively affect the space environment including satellites in low-Earth orbit. Therefore, forecasting these space weather indices is of great importance in space operations and science. In this study, we propose a model based o…
▽ More
Solar radio flux along with geomagnetic indices are important indicators of solar activity and its effects. Extreme solar events such as flares and geomagnetic storms can negatively affect the space environment including satellites in low-Earth orbit. Therefore, forecasting these space weather indices is of great importance in space operations and science. In this study, we propose a model based on long short-term memory neural networks to learn the distribution of time series data with the capability to provide a simultaneous multivariate 27-day forecast of the space weather indices using time series as well as solar image data. We show a 30-40\% improvement of the root mean-square error while including solar image data with time series data compared to using time series data alone. Simple baselines such as a persistence and running average forecasts are also compared with the trained deep neural network models. We also quantify the uncertainty in our prediction using a model ensemble.
△ Less
Submitted 16 December, 2021;
originally announced December 2021.
