Reusable Algorithmic Skeleton Framework for Clustering Algorithms in Wireless Sensor Network (SCW)

Abstract

In Wireless Sensor Networks (WSNs), clustering is often used to improve communication and routing. Therefore, clustering approaches highly attract several researchers since performing clustering saves energy, and energy efficiency is a significant goal in WSN. To beneficially adopt WSN technology, efficient application development is necessary. Therefore, a user-friendly programming abstraction is required to simplify the programming chore without sacrificing efficiency. Using suitable higher-level programming abstraction, it is neither obligatory for a programmer to be an expert in most fields related to WSN nor to be distracted from the application logic by focusing on low-level system issues. To ease the development of new clustering algorithms, a prefabricated algorithmic skeleton, namely SCW, is presented which only requires two functions to be filled in, i.e., to be implemented. The rest of the work (e.g., synchronization, sensing the environment, data aggregation, nodes’ energy calculations, and routing) will be handled by the proposed framework. Hence, SCW has the capability of performing a level of optimization in the background without user interference. By considering software metrics such as Lines of Code (LoC), Halstead metrics, and McCabe complexity while employing the proposed framework, one can implement a WSN clustering algorithm with fewer source lines of code, less programming effort, and difficulty, less time to understand and implement when compared to a built-from-scratch implementation. Although this algorithmic skeleton framework is proposed for implementation, to show its efficiency in this paper, we use the simulation environment.