Location-Aware Caching for Semantic-Based Image Queries in Mobile AD HOC Networks

Mobile image information retrieval, i.e., the processing of mobile image queries, has attracted research attention due to the recent technological advances in mobile and ubiquitous computing, network infrastructures, and multimedia streaming. The previous research focuses on data delivery, while few works have reported on content-based mobile information retrieval. Therefore, it is important to devise effective means to describe the semantics as well as content distribution of mobile data. Caching is an attractive solution that helps reveal semantic relationships among mobile data sources. However, traditional caching techniques rely on exact match of fixed values and are not efficient in dealing with imprecisely described image data. To address these issues, the authors propose a location-aware caching model which reflects the distribution of images based on the analysis of earlier queries. Through extensive simulations, the authors show that the proposed model can perform search with less cost for voluminous data.