Volume 15, Number 1

CFMS: A Cluster-based Convergecast Framework for Dense Multi-Sink Wireless Sensor Networks

  Authors

Gokou Hervé Fabrice Diédié¹, Koigny Fabrice Kouassi² and Tchimou N’Takpé², ¹Université Peleforo Gon Coulibaly, Ivory Coast, ²Université Nangui Abrogoua, Ivory Coast

  Abstract

Convergecast is one of the most challenging tasks in Wireless Sensor Networks (WSNs). Indeed, this data collection process must be conducted while copying with packet collisions, nodes’ congestion or data redundancy. These issues always result in energy waste which is detrimental to network efficiency and lifetime. This paper is aimed to address these problems in large-scale multi-sink WSNs. Inspired by our previous work MSCP, we designed a lightweight protocol stack that seamlessly combines clustering, pathvector routing, sinks’ duty cycling, data aggregation and transmission scheduling in order to minimise message overhead and packet losses. Simulation results show that this solution can mitigate delay while significantly increasing packet delivery and network lifetime.

  Keywords

Wireless Sensor Networks, Multi-sink, Clustering, Convergecast, Aggregation tree, Scheduling.