DASA: An Efficient Data Aggregation Algorithm for LoRa Enabled Fog Layer in Smart Agriculture
Authors :- M Vyas, G Anand, RN Yadav, SK Nayak
Publication :- Advanced Information Networking and Applications. AINA 2023, Lecture Notes in Networks and Systems, Vol 654. Springer, Cham. , 2023.
In a smart agriculture system many resource-constrained sensors are installed near the crops as well as at some strategic locations in an agriculture field to collect relevant crop and environment data in real-time. This data is then used for both critical latency-sensitive decision making as well as for long-term planning. Nowadays, with the help of smart IoT systems, resolving the problems like irrigate fields, avoid animal intrusions, notify the farmer about the seasonal rainfall etc. becomes easier. The edge of the IoT networks regularly receives a huge amount of data generated by sensors that need to be delivered to the server present in the remote data centers/cloud for additional real time control or long term decision making. However, transmitting huge amount of these IoT data across the network toward the cloud imposes a high overhead in terms of bandwidth demand and latency on the IoT network. So, the key challenge in building a smart agriculture system include high communication latency and bandwidth consumption incurred with computing over the data on the cloud. Also, frequent Internet disconnections in rural areas may lead to improper latency sensitive decision making at cloud due unavailability of data. In this paper to resolve such issues of cloud based smart agriculture system, we present a LoRa-based three-tier smart agriculture system comprised of (i) Field layer, (ii) Fog computing layer, and (iii) Cloud computing layer. In particular, a data aggregation algorithm through a LoRa enabled fog computing layer for smart agriculture (DASA) is proposed to compress the total amount of IoT data to be uploaded to the cloud. We present the performance of our proposed scheme and compare with the existing frameworks for smart agriculture system in terms of compression ratio, compression time, compression power, and amount of data transmitted to cloud from fog computing layer. Comparison results show that the proposed algorithm significantly decreases the volume of data to be uploaded to the cloud platform and achieves highest compression ratio among other existing schemes. We also tested the performance of the proposed data aggregation algorithm on real testbed.