Abstract: This paper designs and implements a data sparse-acquisition and transmission system for Wireless Sensor Network (WSN) based on Narrow Band –Internet of Things (NB-IoT) and Field-Programmable Gate Array (FPGA) for smart agriculture. The use of distributed wireless network nodes to collect agricultural environment information in a sparse sampling manner can overcome the disadvantages of limited bandwidth and significantly reduce the amount of data transmission, thereby helping to reduce the energy consumption of wireless network nodes and extend the use time. We use FPGA as the base station to receive the data collected by each node, and make full use of the advantages of parallel computing in FPGA for data recovery. And the method we used are matrix filling algorithm, which was named Latent Factor Model (LFM) in the recommendation system, and gradient descent algorithm. Finally, the recovered complete data will be transmitted to the cloud platform for display in the way of NB-IoT. NB-IoT is an emerging Wide Area Network (WAN) technology that has the advantages of supporting massive connections and being geographically unrestricted. The results show that the data can be perfectly recovered within the specified error range, and the algorithm runs nearly three times faster than the serial operation after adding parallel operations.
Keywords: WSN, FPGA, NB-IoT, LFM
| DOI: 10.17148/IJARCCE.2019.8701