SCADA: scalable cluster-based data aggregation technique for improving network lifetime of wireless sensor networks

To improve the network lifetime of battery-operated wireless sensor networks (WSNs), effective deployment of available low-power resources is required. This is essential as it is impractical to exchange or recharge the energy resource after deployment. To properly utilize the limited energy, a suitable cluster-based data aggregation technique is required, which considerably reduces the network’s overall energy usage. Therefore, this work proposes a cluster-based data aggregation technique called Scalable Clustering Algorithm for Data Aggregation (SCADA). Each cluster has a single cluster head (CH) and several cluster relays (CRs) in this algorithm. The number of CRs in a cluster varies depending on the cluster’s geographic position in relation to the Base Station. We assure that energy usage is reduced as a result of the multi-hop short-distance data transmission by doing so. In hybrid modes, the CHs are also rotated at the appropriate intervals to reduce control packets. The proposed algorithm is novel and well suited for both homogeneous and heterogeneous WSNs because it uses a hybrid CH selection mechanism, a threshold-based CH rotation mechanism, and a mechanism of allocating dedicated CR in each cluster. Effects of variations in the length of the sensing field, variation in SN density, variation in Cluster count, and variation in initial energy and heterogeneity test are used to assess the proposed algorithm’s efficiency. SCADA is incredibly scalable for large-scale WSN and energy-efficient with a small number of control messages as depicted by the analytical results. Simulation-based results, clearly states that as compared to three protocols EESCT, DUCF, and MUCA, SCADA obtains approximately 35%, 25%, and 10%, respectively, more rounds of communication. SCADA drastically improves when the length of the sensing field increases as compared to EESCT, DUCF, and MUCA. SCADA when compared to the EESCT, DUCF, and MUCA algorithms, the average energy consumption is significantly lower by 21 to 38%.