An Empirical Evaluation of BFS, and DFS Search Algorithms on J2ME Platform, and SVG Tiny Parsing on J2ME Platform Using SAX, StAX, and DOM Parsers

Abstract

SVG (Scalable Vector Graphics) Tiny, an XML-based data representation format was used in our Global Train Route Planner J2ME application to render and manipulate train network  images. The SVG Tiny format enables the application to be adaptable with any train network map. We compared three parsing models namely DOM (Document Object Model), SAX (Simple API for XML), and StAX (Streaming API for XML) which were used to visualize the images on mobile phone. We present here the result of the runtime performances, and memory footprints of those parsing models. This is a significant study because handheld devices like mobile phones require seamless interactivity (i.e. high performance) with users and an efficient parsing mechanism with less memory footprints. We also empirically investigated two route searching algorithms - graph and matrix based implementation of DFS (Depth First Search), and matrix based BFS (Breadth First Search) – for performance and memory footprints on a J2ME mobile device emulator. We concluded that DOM parser and DFS based on graph implementation are of better performance than the others.