Semantic overlay network for large-scale spatial information indexing.

Area of Science:

• Computer Science
• Geographic Information Science

Background:

• Growing demand for online spatial information services presents challenges in efficient data indexing within distributed networks.
• Fast indexing of spatial data is crucial for the performance of modern geographic information systems.

Purpose of the Study:

• To propose a novel semantic overlay network, SON_LSII, for efficient large-scale, multi-dimensional spatial information indexing.
• To address the trade-off between indexing efficiency and maintenance overhead in distributed spatial data management.

Main Methods:

• Developed SON_LSII, a hybrid overlay network integrating a semantic quad-tree and Chord ring.
• Implemented a semantic clustering strategy considering data semantics and network performance.
• Utilized a mapping method to reduce multi-dimensional features and an indexing algorithm for complex range queries.

Main Results:

• SON_LSII demonstrates superior scalability, maintenance, indexing hit rates, and adaptability compared to existing overlay networks.
• The network achieves a competitive balance between indexing efficiency and maintenance costs.
• Extensive experiments validate the effectiveness of the proposed indexing approach.

Conclusions:

• The novel SON_LSII network provides an effective solution for large-scale spatial information indexing in distributed environments.
• SON_LSII enhances the performance and manageability of spatial data services.
• This approach is well-suited for handling massive concurrent user access to spatial information.