Ranked Continuous Visible Nearest Neighbor Search

Abstract

Physical obstacles (e.g., buildings, hills, and blindages, etc.) are ubiquitous in the real world, and their existence may affect the visibility between objects and thus the result of spatial queries such as range query, nearest neighbor search, and spatial join, etc. In this paper, we study a novel type of spatial queries, namely, ranked continuous visible nearest neighbor (RCVNN) search, which returns the k visible nearest neighbors that have the maximal optimality according to the predefined optimality metric. We first formulate the problem, and then present an efficient algorithm for RCVNN query processing and prove its correctness. Extensive experimental evaluation demonstrates the efficiency and effectiveness of our proposed algorithm using both real and synthetic datasets.