Cluster Based Data Aggregation Scheme for Latency and Packet Loss Reduction in WSN

• Published in: Computer communications 2020-01, Vol.149, p.36-43
• Main Authors: Seedha Devi, V., Ravi, T., Priya, S. Baghavathi
• Format: Article
• Language: English
• Subjects: Clustering; Compressive sensing; Data aggregation; Slot scheduling; Wireless Sensor Network (WSN)
• ISSN: 0140-3664; 1873-703X
• DOI: 10.1016/j.comcom.2019.10.003

In Wireless Sensor Networks (WSN), the main issues of cluster based data aggregation algorithms are energy balancing, packet loss and latency reduction. In existing scheduling algorithms for data aggregation, time slots are mostly assigned based on the data sensing period and data transmission rate, ignoring packet loss and latency. In this paper, Cluster based Data Aggregation Scheme for Latency and Packet Loss Reduction in WSN is proposed. The proposed scheme consists of two phases: Aggregation Tree Construction and Slot scheduling algorithm. In phase-1, each cluster head applies compressive aggregation for the data received from its members. Then the aggregation tree is constructed by the sink using Minimum Spanning Tree (MST). In phase-2, the packet loss rate and latency are taken into consideration while prioritizing and assigning timeslots to the nodes with aggregated data. This scheme avoids using unnecessary retransmissions and waiting, which results to be beneficial in enhancing the network performance in WSN. Simulation results show that the proposed scheme reduces the latency and overhead and increases the packet delivery ratio and residual energy.