An Internet of Things approach to smart agriculture

It differs from previous research in smart agriculture in that it will build a context-adaptive, secure processing model under strict computation/communication resource and cost constraints, which is vital for widespread adoption by farmers with limited resources and countries that need to employ cost-efficient technologies for smart farming.

Specific scientific research aims that will also have implications for related fields of Internet of Things (IoT) and WSNs within this context are as follows:

1. Creating an extensible framework for IoT-based smart farming that is capable of integrating various types of low power sensors to optimise usage of resources such as water, pesticides and fertilisers under highly variable field and environmental conditions.
2. Devising a low-overhead model of secure communication within WSNs utilised in monitoring field conditions, as well as with the gateway and cloud.
3. Creating an optimal placement and activation strategy for the nodes in a WSN that is adaptable to different contexts for highly reliable and efficient operation, which employs best practices of energy harvesting to achieve an overall long battery life for the devices in the network.
4. Creation of a valuable dataset for machine learning research in agriculture, which includes time-series-based field data and corresponding yield information that will enable smart decision-making in farming applications.

Partners

This project is being carried out in partnership with Middle East Technical University, Turkey.