An extended technique for R-tree to manage multiple type objects

In this paper, we propose a new technique for managing multiple type objects in systems such as GIS and CAD. The novel concept of our mechanism is to separate the objects in a data structure according to the type identifier of the objects. The type identifier is a binary number that signifies a bit pattern corresponding to the type of the object. By means of the type identifier, a data structure can manage not only the space information of objects but also the types of objects. Therefore, in our mechanism, access performance does not depend on the number of all objects, as in most of the past methods, but rather on the number of objects in the type identifier. Also, access and memory performance do not deteriorate even if the types of objects are not independent. The performance of our system is evaluated by simulation tests.