Sampling based δ-approximate data aggregation in sensor equipped IoT networks

Abstract

Emerging needs in data sensing applications result in the usage of IoT networks. These networks are widely deployed and exploited for various efficient data transfer. Wireless sensors can be incorporated in IoT networks to reduce the deployment costs and maintenance costs. One of the critical problems in sensor equipped IoT devices is to design an energy efficient data aggregation method that processes the maximum value query and distinct set query. Therefore, in this paper, we propose two approximate algorithms to process the maximum queries and distinct-set queries in wireless sensor networks. These two algorithms are based on uniform sampling. Solid theoretical proofs are offered which can make sure the proposed algorithms can return correct query results with a given probability. Simulation results show that both δ -approximate maximum value and δ-approximate distinct set algorithms perform significantly better than a simple distributed algorithm in terms of energy consumption.