COCA: Congestion-oriented clustering algorithm for wireless sensor networks

• Published in: 2016 8th IEEE International Conference on Communication Software and Networks (ICCSN) pp. 450 - 454
• Main Authors: Khanh Le, Tho Quan, Thang Bui, Petrucci, Laure
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2016
• Subjects: Algorithm design and analysis; Clustering algorithms; congestion detection; Measurement; Network topology; Sensors; Topology; Wireless sensor networks
• DOI: 10.1109/ICCSN.2016.7587199

Wireless Sensor Networks (WSNs) are used for a lot of monitoring applications on risky area such as volcanoes, earthquake zones and deep jungle regions. Such network topologies are randomly deployed by dropping large sets of sensors from helicopters. Thus, one cannot ensure that the actual post-deployment topology of sensors is reliable, i.e. whether it suffers from possible congestion or not, as the congestion detection on a WSN is by no means a trivial task. It is even worse when the number of sensors deployed in real-life situation is very large. Several clustering techniques have been reported to handle large-scale WSNs, but none of which specifically addresses the congestion issues. In this paper, we discuss an effective clustering-based approach for congestion detection on WSN, known as COCA (Congestion-Oriented Clustering Algorithm). This algorithm groups sensors with a high chance of congestion into clusters based on distance or imbalance metrics, which are standard metrics for congestion evaluation in WSNs. Thus, COCA enjoys promising results in our experiments, as compared to existing approaches.