Surrogate Models for Spatiotemporal Data Reconstruction

The patent creates surrogate models as simplified copies of the original spatiotemporal datasets. These surrogate models capture the essential patterns and relationships of the data without requiring distribution of the full dataset. The surrogate models can be distributed to end users, who can then locally reconstruct or approximate the original data, significantly reducing the volume of data that needs to be stored and transmitted while maintaining data access capability.

The patent extracts the essential patterns, relationships, and key information from the full spatiotemporal datasets to create compact surrogate models. These surrogate models contain only the critical information needed to reconstruct or approximate the original data, eliminating redundant or less important data elements. This extraction process maintains sufficient data fidelity for most applications while dramatically reducing the amount of data that needs to be transmitted over networks.

The patent enables end users to reconstruct or approximate high-resolution spatiotemporal data locally on their own devices using the distributed surrogate models. Instead of providing all users with identical high-resolution datasets (which would be expensive to distribute), each user can generate the data they need locally based on their specific analysis requirements. This approach maintains analysis accuracy where needed while significantly reducing distribution costs.