A Novel Power-Efficient Data Aggregation Scheme for Cloud-Based Sensor Networks

1. Introduction

Wireless sensor networks as the name suggests can be defined as the network of several wireless entities primarily various sensors (Roy & Chandra, 2019). Sensors prove as of great utility for determining real-time physical changes as they can pick up changes in the surroundings and update the data related to it through which further recommendations can be justified. Wireless sensor networks have found massive number of applications in recent times. They are used in area monitoring, habitat monitoring, health monitoring, smart cities (Giliberto et al., 2019) etc. The sensor nodes could be deployed effortlessly in difficult terrains and Warfield for military assistance. Also the cost to setup wireless sensor network is not much. All these factors have contributed in the rapid technological advancements in the area of wireless sensor networks. The sensor nodes deployed to monitor environment, industrial machinery etc. in real time can produce enormous amount of data.(Arena & Pau, 2020) But the sensors deployed are very fragile and have very less computational power. So energy conservation in such type of networks becomes very crucial. The major problem of the network is the limited battery. Higher battery usage causes less network lifetime which may lead to crashing down the network and hardware. Therefore, to provide the longest uptime of the network, the battery issue is to be settled with utmost priority. Various research studies suggested event-based utilization of the sensor motes (Padmaja & Marutheswar, 2018; Zhang et al., 2020). Sleep and wake as the initial method used to control the power consumption of the motes gradually became less feasible and more complex. The sensor motes are expected to finally transmit aggregated data to the sink node to process. Overhead data transmission causes the battery drainage at a much faster pace depending upon the number of hops. Considering the ultimate destination of the useful transmission as sink node it can be worked upon. Although the sink node is privileged to have power in abundance same cannot be assigned to all sensors as cost-measure. Therefore, conserving data aggregation schemes with minimal transmissions to the sink node can be deployed. Some important aspects of this scheme are non-redundancy and complete data aggregation in data collection. Redundancy can be mitigated by efficient data-aggregation techniques as it directly affects the energy requirements.