A wireless data clustering method for multipoint queries

Abstract

By effective data clustering, the mobile clients can access the data objects of their interest in short latency. In this paper, we propose a clustering method of wireless data for multipoint queries, where wireless data objects are uniformly broadcasted. We first analyze the data clustering problem and propose a new measure, named the Query Distance, i.e., convenient to manipulate for the analyses of the average query performance. Then, we propose a clustering method that efficiently minimizes the Query Distance, based on the Gray coding scheme. We show that the Gray coding scheme has useful properties that can be utilized for clustering the data on the air. We also show the performance advantage of our method through experiments. Since, as far as we are aware of, there is no previous work that deals with multipoint queries in wireless data broadcasting, the experiments are in comparison with the random clustering method.