Abstract:
With large advances in the wireless network technologies, internet, and mobile computing devices such as smart phones that enable users to access latest updated information ubiquitously from internet in an "anytime, anywhere, and any cost" style on their mobile phone devices. However, in this model user always require internet connectivity such as data or Wi-Fi connection to access updated information. But, the most common challenge is that data connection is typically more expensive, Wi-Fi connection is usually not available everywhere, and often it is not even free for all users.
To access updated information on mobile devices at anytime and anywhere regardless of network connectivity with cheapest prices or even free of cost, a good option can be to deliver latest information from web server to users through short message service (SMS). The main idea is to utilize short message service to deliver information due to its pervasive availability and worldwide support by various mobile devices and wireless cellular networks. The primary goal of our research is to design and develop a geographically distributed algorithm for SMS based applications that delivers the updated information contents from source (web server) to various mobile subscribers based on their current geographical location via SMS. So we have proposed a geographical location-based dynamic hierarchical clustering (DHC) algorithm for grouping large number of cellular subscribers and head node selection algorithm for selecting some of the user's mobile devices as network relay nodes. We have used users' interest with their current geographical vicinity to make clusters while considered device capacity for selecting relay nodes. The performance and efficiency of proposed algorithms is evaluated and validated using various parameters by simulating it on Google maps using JavaScript and Google maps API. At the end we have developed a prototype system. This prototype system validated our design and proposed algorithms, and also evaluated the performance of proposed algorithms.
