Wireless Sensor Network (WSN) based smart irrigation solutions rely on sensors for data reporting, a central infrastructure for generating irrigation parameters and a mobile machinery called Linear for irrigating. In this scenario, change in soil and environmental properties is referred as an event. Spatial distribution of such properties leads to multiple event regions in farmland. Identifying Boundary Nodes (BNs) of an event region is a well investigated problem in WSN, however, data reporting from BNs that caters to application specific requirements have rarely studied. To this end, an energy efficient data harvesting protocol for smart irrigation is designed that integrates three modules. First, to reduce number of source nodes and hence, energy consumption, a distributed Representative Node (RN) selection mechanism is proposed to identify a subset of BNs as the source nodes. Information from these RNs ensures correct actions even when nodes of more than two event regions lie nearby that is not achieved by existing RN selection mechanisms. Second, stop time of Linear must be adequate to collect data from RNs for ensuring proper actions. However, existing mechanisms limit to optimizing network lifetime. Calculation of data collection dependent stop time of Linear without having any information of RNs is proposed. Third, Reduced overhead broadcast scheme is proposed that uses single message transmission by a node for Linear location information dissemination. Results indicate that at least 58% of BNs are required for accurate region construction and overall, 12% to 56% less energy is consumed in the data harvesting process.
Jha, S., Trivedi, A., Pattanaik, K.K., Gauttam, H., Bellavista, P. (2024). Efficient data harvesting from boundary nodes for smart irrigation. PEER-TO-PEER NETWORKING AND APPLICATIONS, 17(1), 139-162 [10.1007/s12083-023-01577-8].
Efficient data harvesting from boundary nodes for smart irrigation
Bellavista, Paolo
2024
Abstract
Wireless Sensor Network (WSN) based smart irrigation solutions rely on sensors for data reporting, a central infrastructure for generating irrigation parameters and a mobile machinery called Linear for irrigating. In this scenario, change in soil and environmental properties is referred as an event. Spatial distribution of such properties leads to multiple event regions in farmland. Identifying Boundary Nodes (BNs) of an event region is a well investigated problem in WSN, however, data reporting from BNs that caters to application specific requirements have rarely studied. To this end, an energy efficient data harvesting protocol for smart irrigation is designed that integrates three modules. First, to reduce number of source nodes and hence, energy consumption, a distributed Representative Node (RN) selection mechanism is proposed to identify a subset of BNs as the source nodes. Information from these RNs ensures correct actions even when nodes of more than two event regions lie nearby that is not achieved by existing RN selection mechanisms. Second, stop time of Linear must be adequate to collect data from RNs for ensuring proper actions. However, existing mechanisms limit to optimizing network lifetime. Calculation of data collection dependent stop time of Linear without having any information of RNs is proposed. Third, Reduced overhead broadcast scheme is proposed that uses single message transmission by a node for Linear location information dissemination. Results indicate that at least 58% of BNs are required for accurate region construction and overall, 12% to 56% less energy is consumed in the data harvesting process.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.