Statistical Link Acquisition Estimation in Satellite Constellations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In satellite constellations, especially LEO satellite networks, inter-satellite link acquisition times are unpredictable and time-consuming due to relative satellite motion, leading to impactful network outages, as current methods fail to accurately estimate when links will be re-established, resulting in unnecessary delays and potential routing issues.
Innovation Solution
A statistical-based computation method and apparatus that estimate inter-satellite link acquisition times by determining a statistical model based on prior link acquisitions, allowing for the dissemination of these estimates throughout the network, enabling more accurate prediction of link availability and reducing network downtime.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conservative link acquisition time estimates are used, then network reliability is improved by avoiding premature link declarations, but network productivity deteriorates due to unnecessary delays and reduced throughput
Solution Approach 1:
The system continuously monitors actual link acquisition times and uses this feedback to refine and update the statistical models. This allows the network to learn from past performance and improve its estimation accuracy over time, resolving the contradiction between reliability and productivity by adapting to actual conditions rather than relying on fixed conservative estimates
Solution Approach 2:
The patent changes the parameter used for link acquisition time estimation from fixed conservative values to dynamic statistical parameters (mean, variance, confidence intervals) that are continuously updated based on observed data. This allows the system to optimize the balance between reliability and productivity by adjusting estimates based on actual network conditions
2Measurement precision
If statistical models are disseminated throughout the satellite constellation, then measurement precision of link acquisition times is improved, but device complexity increases due to additional computation and communication requirements
Solution Approach 1:
The system segments the complex statistical modeling task by having each satellite maintain only its own local statistical model for link acquisitions involving itself, while disseminating these models to neighbors. This divides the computational burden across multiple nodes rather than requiring each node to maintain comprehensive global models, reducing individual device complexity while improving overall measurement precision
Solution Approach 2:
Each satellite autonomously maintains and updates its own statistical model based on observed link acquisition data, without requiring centralized control or complex coordination. This self-service approach allows satellites to independently perform precise measurements and share their insights, improving measurement precision while keeping individual device complexity manageable
Data Source
AI summary
The present invention provides a method and apparatus for supporting estimation of inter-satellite link acquisition times in a satellite constellation. The method includes computing or generating an indication of a statistical model based on observations for prior link acquisition times. The method further includes communicating an indication such as a statistical model for link acquisition times or related parameters through a communication network, or a combination thereof. The indication may be communicated using one or more transmission techniques or protocols, such as flooding, a link state protocol or gossip protocol. Based on the disseminated indication, future link acquisition times can be predicted by satellites in the satellite constellation. Embodiments of the invention use a statistical-based computation approach, such as regression modelling or random variable modelling, to estimate link acquisition times or associated estimation parameters. The estimates or associated estimation parameters may then be disseminated through the constellation.